Issue 16526: Definition of proposition (sbvr-rtf) Source: NIST (Mr. Edward J. Barkmeyer, edbark(at)nist.gov) Nature: Uncategorized Issue Severity: Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Resolution: Revised Text: Actions taken: August 31, 2011: received issue Discussion: End of Annotations:===== te: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) From: "Donald Chapin" To: Subject: RE: issue 16526 -- SBVR RTF issue -- An Alternative Solution Date: Fri, 2 Sep 2011 15:17:57 +0100 X-Mailer: Microsoft Outlook 14.0 Thread-Index: Acxpex00uAt0INQHRFuyqz5FCmJ1/Q== X-Mirapoint-IP-Reputation: reputation=Fair-1, source=Queried, refid=tid=0001.0A0B0303.4E60E597.0090, actions=TAG X-Junkmail-Premium-Raw: score=7/50, refid=2.7.2:2011.7.19.51514:17:7.944, ip=81.149.51.65, rules=__TO_MALFORMED_2, __TO_NO_NAME, __BOUNCE_CHALLENGE_SUBJ, __BOUNCE_NDR_SUBJ_EXEMPT, __HAS_MSGID, __SANE_MSGID, __MIME_VERSION, __CT, __CTYPE_MULTIPART_ALT, __CTYPE_HAS_BOUNDARY, __CTYPE_MULTIPART, __HAS_X_MAILER, __OUTLOOK_MUA_1, __USER_AGENT_MS_GENERIC, __ANY_URI, __FRAUD_CONTACT_NUM, __STOCK_PHRASE_24, __CP_URI_IN_BODY, __C230066_P5, __CP_NOT_1, __FRAUD_CONTACT_NAME, __STOCK_PHRASE_7, __HTML_MSWORD, __HTML_BOLD, __HTML_FONT_BLUE, __HAS_HTML, BODY_SIZE_10000_PLUS, __MIME_HTML, __TAG_EXISTS_HTML, __STYLE_RATWARE_2, RDNS_GENERIC_POOLED, HTML_90_100, RDNS_SUSP_GENERIC, __OUTLOOK_MUA, RDNS_SUSP, FORGED_MUA_OUTLOOK X-Junkmail-Status: score=10/50, host=c2bthomr07.btconnect.com X-Junkmail-Signature-Raw: score=unknown, refid=str=0001.0A0B0205.4E60E60D.01CD,ss=1,fgs=0, ip=0.0.0.0, so=2010-07-22 22:03:31, dmn=2009-09-10 00:05:08, mode=multiengine X-Junkmail-IWF: false An Alternative Solution In place of current .statement.: proposition statement Definition: sentence that has only one reasonable interpretation in the grammar of the natural language in which it is stated and that is either true or false in any given possible world In place of current .proposition.: proposition meaning Definition: meaning of a given proposition statement or (two or more) semantically equivalent proposition statements Donald From: Juergen Boldt [mailto:juergen@omg.org] Sent: 01 September 2011 15:11 To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org Date: Fri, 02 Sep 2011 14:08:47 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: Donald Chapin CC: "sbvr-rtf@omg.org" Subject: Re: issue 16526 -- SBVR RTF issue -- An Alternative Solution X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p82I8qZa014770 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315591734.41695@vgj/CGoN61XVYGzTNhTknw X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Donald Chapin wrote: _An Alternative Solution_ In place of current .statement.: proposition statement Definition: sentence that has only one reasonable interpretation in the grammar of the natural language in which it is stated and that is either true or false in any given possible world In place of current .proposition.: proposition meaning Definition: meaning of a given proposition statement or (two or more) semantically equivalent proposition statements This seems to do a lot of violence to the current specification. Currently: A 'statement' is the representation of a proposition by an expression (that is a "sentence" in some grammar). A 'proposition' 'is true' or 'is false'. Donald's proposal undefines 'statement' and appears to make it mean the 'expression', rather than the 'representation'. Further, it apparently creates 'proposition statement is true' and 'proposition statement is false', which assigns the truth value to the expression, rather than the meaning. Finally, it reverts to the original SBVR Alpha text in which 'proposition' is defined as the meaning of some expression, without having any clear characterization of the expression. At least the current text provides a unique characteristic of the meaning that distinguishes it from concepts. In any case, Donald's proposal moves away from what both Date/Time and Don's possible states of affairs need -- the idea that a proposition is a meaning that conceptualizes situations, regardless of whether it is true or false. The nature of the semantic structure of a 'sentence', and thus of the proposition as a meaning, is that it conceptualizes a situation in terms of specific (or quantified) things playing specific roles in a relationship or activity that is characterized by verb concepts. The current SBVR fact type 'state of affairs involves thing in role' is more true of the proposition than of the state itself. That is, if we take reality to exist beyond each individual perception of it, then the same real situation can have different interpretations in the minds of different observers. But any individual observer conceptualizes the situation in terms of particular verb concepts and things playing their fact type roles. And a 'sentence' formalizes that conceptualization using grammatical verbs for the verb concepts and grammatical role assignments for grammatical referencers for the things. So it is much easier to define 'sentence' as a chosen form of expression for the conceptualization (the 'proposition'). Of course, SBVR should not try to distinguish between perception and reality, (or between belief, supposition and actuality, for that matter). The idea here is only to get the cart and horse in the right order. In real world encounters, the sequence is: happening, observation, conceptualization, expression In planning and risk analysis and so on, the sequence is: experience (prior observations and conceptualizations), analogy, conceptualization, expression In communication, the pattern is: speaker: ..., conceptualization, expression, listener: interpretation, conceptualization In all cases, however, the conceptualization precedes the expression, that is, the proposition exists before the sentence. In the mind of the listener, however, the interaction with the sentence is interpretation, not expression, and for him/her the sentence apparently precedes the conceptualization.. It is easy to see oneself and one's tooling as the reader/listener instead of the author/speaker, and that gives rise to defining the meaning in terms of the expression artefact, but the goal of SBVR is to support effective expression of business intent, not improved interpretation of existing expression forms. -Ed Donald *From:* Juergen Boldt [mailto:juergen@omg.org] *Sent:* 01 September 2011 15:11 *To:* issues@omg.org; sbvr-rtf@omg.org *Subject:* issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer > Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF > Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org [] -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." From: Don Baisley To: "sbvr-rtf@omg.org" Subject: RE: issue 16526 -- SBVR RTF issue Thread-Topic: issue 16526 -- SBVR RTF issue Thread-Index: AQHMaLFXF0y3f6q7s0yD7Sl4Xh8oHJVEozmA Date: Fri, 9 Sep 2011 07:26:50 +0000 Accept-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [157.54.51.34] Here are some ideas regarding issue 16526. As the issue points out, a proposition corresponds to a state of affairs and can be used to refer to that state of affairs without regard to whether the state of affairs is actual. The truth of the proposition is based whether the state of affairs is actual. Since a state of affairs is either actual or not actual, the proposition is either true or false, regardless of whether its truth or falseness is of interest or is known. There does not appear to be a reason to change the definition of .proposition.. The issue seems to be about nested propositions. E.g., in the formulation of the rule, .It is prohibited that a person drives while drunk., there is a closed formulation of the proposition, .a person drives while drunk.. That proposition corresponds to the state of affairs that is prohibited by the rule. If that proposition is true, then there is a violation of the rule. The truth of that proposition is independent of the truth of the rule, which is true if there truly is such a prohibition. Another example involves tense: .EU-Rent has been in debt.. The formulation of that proposition embeds a closed formulation of another proposition, .EU-Rent is in debt.. That closed formulation is objectified so that the outer formulation can refer to EU-Rent being in debt as a situation that .has happened.. Time is in the universe of discourse, so rather than relating propositions to time, which makes a messy crossover between meanings and the universe of discourse, it is best practice to relate states of affairs to time, because the states of affairs are in the universe of discourse along with time. The instances of their being related are in the universe of discourse too. So we don.t say a proposition has been true, is false now and will be true again. Rather, we relate the corresponding state of affairs to time (it has happened). The proposed Date-Time submission provides fact types to relate time to some states of affairs . those that are called .occurrences.. It does not provide for relating states of affairs generally to time. For example, it does not provide a characteristic such as .state of affairs has happened. which is needed to formulate the example above. Since SBVR submission uses time concepts in its examples, and since the Date-Time submitters have decided to not provide fact types to relate states of affairs generally to time, SBVR could add some basic fact types that build on the Date-Time concepts of past and future. Here are two: state of affairs has happened (a.k.a. .has been actual.) state of affairs is going to happen (a.k.a. .is going to be actual.) These fact types could be added formally in 8.6 or as examples for the purpose of helping to explain transient states of affairs, which are states of affairs that are actual at some times without being actual at all times. Regards, Don From: Juergen Boldt [mailto:juergen@omg.org] Sent: Thursday, September 01, 2011 7:11 AM To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org Date: Fri, 09 Sep 2011 11:41:22 +0100 From: John Hall User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20110812 Thunderbird/6.0 To: edbark@nist.gov CC: Donald Chapin , "sbvr-rtf@omg.org" Subject: Re: issue 16526 -- SBVR RTF issue -- An Alternative Solution X-Mailcore-Auth: 4600872 X-Mailcore-Domain: 13170 On 02/09/2011 19:08, Ed Barkmeyer wrote: ... Of course, SBVR should not try to distinguish between perception and reality, (or between belief, supposition and actuality, for that matter). The idea here is only to get the cart and horse in the right order. In real world encounters, the sequence is: happening, observation, conceptualization, expression In planning and risk analysis and so on, the sequence is: experience (prior observations and conceptualizations), analogy, conceptualization, expression In communication, the pattern is: speaker: ..., conceptualization, expression, listener: interpretation, conceptualization In all cases, however, the conceptualization precedes the expression, that is, the proposition exists before the sentence. In the mind of the listener, however, the interaction with the sentence is interpretation, not expression, and for him/her the sentence apparently precedes the conceptualization.. It is easy to see oneself and one's tooling as the reader/listener instead of the author/speaker, and that gives rise to defining the meaning in terms of the expression artefact, but the goal of SBVR is to support effective expression of business intent, not improved interpretation of existing expression forms. Ed, I agree with what you wrote if we are concerned only with individuals, thinking independently. But we use SBVR to develop shared vocabularies and rules for businesses (or other organizations or groups of people). The people in a business who manage its vocabulary and rules have to get to consensus, which is then the 'official' version, that provides: business definitions used in legal contracts, marketing and advertising, product and service specifications, job descriptions, compliance reporting, etc. the basis for developing data models for information systems, and business semantics for their content People can get to consensus only by agreeing on the concepts and guidance, and they do that by discussing written definitions. Frequently, individuals will change their minds during the discussion. I don't mean that the concepts change - rather that some individuals recognize that some concepts they, as individuals, started with should be replaced by other concepts that take account of other people's views, and better fit the needs of the business as a whole. There has recently been quite a lot of discussion towards consensus in the SBVR RTF and the Date-Time submission team. So, for developing shared vocabulary and rules, I would extend your 'real world encounters' sequence as: happening, observation, individual conceptualization, individual expression, participation in reaching consensus on expression, assimilation of shared conceptualization This is related to your sequence for communication. The problem there, in developing shared vocabulary and rules, is when people in the business make different interpretations, so that there is consensus only on the expression and not on the meaning, and people don't act consistently in the business. The major safeguard against this is the quality and precision of the expressions. Regards, John To: sbvr-rtf@omg.org Cc: date-time@omg.org Subject: RE: issue 16526 -- SBVR RTF issue X-KeepSent: 99CA7F02:FF31262F-85257906:0040FB21; type=4; name=$KeepSent X-Mailer: Lotus Notes Release 8.5.1FP5 SHF29 November 12, 2010 From: Mark H Linehan Date: Fri, 9 Sep 2011 08:38:26 -0400 X-MIMETrack: Serialize by Router on D01MC604/01/M/IBM(Release 8.5.2FP1 ZX852FP1HF6|May 2, 2011) at 09/09/2011 08:38:28, Serialize complete at 09/09/2011 08:38:28 Don misunderstands the Date-Time vocabulary when he says: > The proposed Date-Time submission provides fact types to relate time > to some states of affairs . those that are called âoccurrencesâ. &nbbsp; > It does not provide for relating states of affairs generally to > time. For example, it does not provide a characteristic such as â > state of affairs has happenedâ which is needed to formulate the > example above. Since SBVR submission uses time concepts in its > examples, and since the Date-Time submitters have decided to not > provide fact types to relate states of affairs generally to time, > SBVR could add some basic fact types that build on the Date-Time > concepts of past and future. Here are two: > > state of affairs has happened (a.k.a. âhas been actualâ) > state of affairs is going to happen (a.k.a. âis going to be actualâ) > > These fact types could be added formally in 8.6 or as examples for > the purpose of helping to explain transient states of affairs, which > are states of affairs that are actual at some times without being > actual at all times. > Date-Time has avoided using the term "state of affairs" at the request of the SBVR RTF. That does not mean that Date-Time does not have fact types equivalent to what Don requests. Date-Time uses the term "situation model" for "relating states of affairs generally to time." Clause 14.3 of Date-Time provides these fact types that are relevant to Don's proposal: situation model is in the past situation model is in the future situation model is accomplished situation model is continuous occurrence exemplifies situation model Date-Time Table 2 on page 95 shows 12 ways combinations of these characteristics. Here's one way to put these together to achieve the meaning of Don's 'state of affairs has happened (a.k.a. âhas been actualâ)'. To start, I note that âhas been actualâ uses what English calls the present tense, progressive and perfect aspects. "Progressive" aspect is the sense that a a situation applies over a period of time. "Perfect" aspect means that the situation has completed, i.e. reached an end. It does not mean that the situation has "happened". I start with an example situation model, "EU-Rent being profitable". The present tense, continuous and perfect aspect of this is "EU-Rent has been profitable". I believe this is a reasonable interpretation of what Don means by "state of affairs has happened", where "has happened" is understood as âhas been actualâ. I would formulate it as follows: exists s:situation model unitary where and "proposition describes situation model"('EU-Rent being profitable', s) and "situation model is accomplished" "situation model is continuous" exists o:occurrence where "occurrence exemplifies situation model"(o, s) If Don intended "state of affairs has happened" to be past tense then I would add "situation model is in the past" to the formulation. Date-Time would apply a similar formulation for 'state of affairs is going to happen (a.k.a. âis going to be actualâ'. The formulation would add "state of affairs is in the future". What Don is suggesting for clause 8.6 is a subset of what is really needed, and a subset that overlaps Date-Time. I suggest that it would be a mistake for SBVR to include an incomplete and inadequate subset of the real needs of modeling the time aspects of language. -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research From: Don Baisley To: "sbvr-rtf@omg.org" Date: 09/09/2011 03:28 AM Subject: RE: issue 16526 -- SBVR RTF issue -------------------------------------------------------------------------------- Here are some ideas regarding issue 16526. As the issue points out, a proposition corresponds to a state of affairs and can be used to refer to that state of affairs without regard to whether the state of affairs is actual. The truth of the proposition is based whether the state of affairs is actual. Since a state of affairs is either actual or not actual, the proposition is either true or false, regardless of whether its truth or falseness is of interest or is known. There does not appear to be a reason to change the definition of âpropositionâ. The issue seems to be about nested propositions. E.g., in the formulation of the rule, âIt is prohibited that a person drives while drunkâ, there is a closed formulation of the proposition, âa person drives while drunkâ. That proposition corresponds to the state of affairs that is prohibited by the rule. If that proposition is true, then there is a violation of the rule. The truth of that proposition is independent of the truth of the rule, which is true if there truly is such a prohibition. Another example involves tense: âEU-Rent has been in debt.â The formulation of that proposition embeds a closed formulation of another proposition, âEU-Rent is in debtâ. That closed formulation is objectified so that the outer formulation can refer to EU-Rent being in debt as a situation that âhas happenedâ. Time is in the universe of discourse, so rather than relating propositions to time, which makes a messy crossover between meanings and the universe of discourse, it is best practice to relate states of affairs to time, because the states of affairs are in the universe of discourse along with time. The instances of their being related are in the universe of discourse too. So we donât say a proposition has been true, is false now and will be true again. Rather, we relate the corresponding state of affairs to time (it has happened). The proposed Date-Time submission provides fact types to relate time to some states of affairs . those that are called âoccurrencesâ. &nnbsp; It does not provide for relating states of affairs generally to time. For example, it does not provide a characteristic such as âstate of affairs has happenedâ which is needed to formulate the example above. Since SBVR submission uses time concepts in its examples, and since the Date-Time submitters have decided to not provide fact types to relate states of affairs generally to time, SBVR could add some basic fact types that build on the Date-Time concepts of past and future. Here are two: state of affairs has happened (a.k.a. âhas been actualâ) state of affairs is going to happen (a.k.a. âis going to be actualâ) These fact types could be added formally in 8.6 or as examples for the purpose of helping to explain transient states of affairs, which are states of affairs that are actual at some times without being actual at all times. Regards, Don From: Juergen Boldt [mailto:juergen@omg.org] Sent: Thursday, September 01, 2011 7:11 AM To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org Date: Fri, 09 Sep 2011 13:07:40 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: Don Baisley CC: "sbvr-rtf@omg.org" Subject: Re: issue 16526 -- SBVR RTF issue X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p89H7jNW027413 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1316192866.58442@DcMZvVMgLL6gWXvsUYZATQ X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Don Baisley wrote: Here are some ideas regarding issue 16526. As the issue points out, a proposition corresponds to a state of affairs and can be used to refer to that state of affairs without regard to whether the state of affairs is actual. The truth of the proposition is based whether the state of affairs is actual. Since a state of affairs is either actual or not actual, the proposition is either true or false, regardless of whether its truth or falseness is of interest or is known. Given a fact model comprising: John is in London during 5-8 September 2011. John is in St. Imier during 9-10 September 2011. The time of interest in the Universe of Discourse is Sept 5-10, 2011. The proposition expressed by 'John is in London' is used in representing this fact model, both in SBVR examples, and formally in Date/Time. Is it true or false? I believe that in that actual world, the proposition is neither true nor false. Therefore, it is a counterexample to what Don says. There does not appear to be a reason to change the definition of .proposition.. According to the SBVR definition, in that actual world, the expression 'John is in London' does not represent a proposition, since it is neither true nor false. Therefore, a logical formulation of that expression cannot be the 'logical formulation' of an 'objectification', and the SBVR example of using 'John is in London' in an objectification in formulating 'John is in London during 5-8 September 2011.' is not valid. Therefore, there is a reason to change the definition of 'proposition'. The issue seems to be about nested propositions. E.g., in the formulation of the rule, .It is prohibited that a person drives while drunk., there is a closed formulation of the proposition, .a person drives while drunk.. That proposition corresponds to the state of affairs that is prohibited by the rule. If that proposition is true, then there is a violation of the rule. The truth of that proposition is independent of the truth of the rule, which is true if there truly is such a prohibition. No that is not an example of the problem, and Don knows this full well. This is a diversion. Look at the example above, or the one below. Another example involves tense: .EU-Rent has been in debt.. Which is not a concept in SBVR, but it is defined in Date/Time. The formulation of that proposition embeds a closed formulation of another proposition, .EU-Rent is in debt.. That closed formulation is objectified so that the outer formulation can refer to EU-Rent being in debt as a situation that .has happened.. Time is in the universe of discourse, so rather than relating propositions to time, which makes a messy crossover between meanings and the universe of discourse, it is best practice to relate states of affairs to time, because the states of affairs are in the universe of discourse along with time. The instances of their being related are in the universe of discourse too. So we don.t say a proposition has been true, is false now and will be true again. Rather, we relate the corresponding state of affairs to time (it has happened). I see. Time creates a problem for the SBVR definition of 'proposition' (which is what the issue says). So let us claim that the issue is addressing the wrong question. We should be talking about the "corresponding state of affairs". Problem: There cannot BE a "corresponding state of affairs" if 'EU-Rent is in debt' does not represent a proposition. If the meaning of 'EU-Rent is in debt' is not either true or false, it is not a proposition according to the SBVR definition of 'proposition'. A meaningless sentence does not 'correspond to' anything. That is the issue! The following is irrelevant to the issue, because the problem exists in SBVR v1.1interim, without regard to Date/Time: The proposed Date-Time submission provides fact types to relate time to some states of affairs ­ those that are called .occurrences.. It does not provide for relating states of affairs generally to time. For example, it does not provide a characteristic such as ._state of affairs_ has happened. which is needed to formulate the example above. Since SBVR submission uses time concepts in its examples, and (It is, BTW, also totally inaccurate. Date/time provides a vocabulary in which these propositions can be formulated, in clause 14.3.) The following suggestion is simply outrageous: since the Date-Time submitters have decided to not provide fact types to relate states of affairs generally to time, SBVR could add some basic fact types that build on the Date-Time concepts of past and future. Here are two: _state of affairs_ has happened (a.k.a. .has been actual.) _state of affairs_ is going to happen (a.k.a. .is going to be actual.) These fact types could be added formally in 8.6 or as examples for the purpose of helping to explain transient states of affairs, which are states of affairs that are actual at some times without being actual at all times. It would be a violation of the scope of SBVR as adopted for the RTF to add a model of a concept that was not in the specification. It is an egregious violation of the scope of the RTF to do so in the presence of an outstanding RFP with proposal for addressing exactly this domain. If Microsoft intended to propose such a thing, it had only to issue a Letter Of Intent during the time period appointed by the OMG RFP. This is merely flippant. -Ed Regards, Don *From:* Juergen Boldt [mailto:juergen@omg.org] *Sent:* Thursday, September 01, 2011 7:11 AM *To:* issues@omg.org; sbvr-rtf@omg.org *Subject:* issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org [] -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." To: sbvr-rtf@omg.org Cc: date-time@omg.org Subject: meaning of 'proposition' X-KeepSent: 4462B8F5:8E2F3A1F-8525790D:005B17B6; type=4; name=$KeepSent X-Mailer: Lotus Notes Release 8.5.1FP5 SHF29 November 12, 2010 From: Mark H Linehan Date: Fri, 16 Sep 2011 12:46:17 -0400 X-MIMETrack: Serialize by Router on D01MC604/01/M/IBM(Release 8.5.2FP1 ZX852FP1HF6|May 2, 2011) at 09/16/2011 12:46:20, Serialize complete at 09/16/2011 12:46:20 During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "Re: issue 16526 -- SBVR RTF issue". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research Subject: Re: meaning of 'proposition' From: keri Date: Fri, 16 Sep 2011 10:59:35 -0700 Cc: sbvr-rtf@omg.org, date-time@omg.org To: Mark H Linehan X-Mailer: Apple Mail (2.1084) Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, independent of knowing any facts. I also captured points such as: Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is based on the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is based on. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "Re: issue 16526 -- SBVR RTF issue". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research Date: Fri, 16 Sep 2011 16:46:21 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: Stan Hendryx CC: "sbvr-rtf@omg.org" , "date-time@omg.org" Subject: Re: meaning of 'proposition' X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p8GKkQYO014512 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1316810789.2307@NULZb6tQ2XXe+6KB1YGhHA X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? -Ed Stan On Sep 16, 2011, at 10:59 AM, keri > wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." X-Yahoo-Newman-Id: 705029.5647.bm@omp1030.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316210857; bh=8XzrGNzypw2dQ4a0QgFT+uC5jeD4U6e8fSzTzaRnxDA=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type; b=jFMmWTwzDz6x301XdPs8WcL53OY+Ta1Vw6930ek040uefLsb9dyX3JuQRCQyk0uBPdnJEViavcq0ycGXUxK4ncelNGJXiBsHYRl+vtDM2pFTR1I2WA5rtp3y2D9hMXjgmXOCm9XjqfzK953CXfSJYWCC1o4GFKtL6YJyBSJMlJs= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: wQcIvuQVM1kyn2XlBuEPH7abVGGI9D2KVGeXQVtyIYoOePJ cnE.1EApns8P3wq8joYo5doQqS7EQARbwabW0gX7.Rj349htETbxAsfGFYf9 54zrgoMf_QEQTfAZlmZs1wXJUL8MeRjflMZOwKKxjNF.ej30I3m6YL.xZ4S5 vvV5KzHYoBEC1Ufp.2ZwdIj3Syu9p1a.7ZHSanyJGg.ogWjIjSc1jvey4bE4 zxWpPunHZ87x.PpBph.ZDVRyoWjPXI1XVy38r9jWIgn.POuFcjQj0kYs7hJe Opw60Qx4xLQJipJUWWc75WcoCvvS7ym27gWE2iJQFiWvlgxe98GDvGRunWYX CfkJIMTUJJasqpFD9ecd7YLquy_vapDWYT45TXX82wbTTqJAORTiHwRmwDTE SGZLJKIZAhwTN2grGxGBvnIle.8rb.W0nxda1TC7BZUFrASkDpmPG.4djLJT whEgVzTnJQX5Q1RMB8_avrJgg1UaG7HouaXq8OhX4FcvA2Uv8BOVM_wzK32s uMP43asH_G32q.H.Dk30kr4tOMtfWedwfEUCMJYaJDsSUEkyo7icKNkV_u4H NwtlgMfb2XqMkRNWl.c0GniASTOhRkPL8CKAgi7z_73hDjblsRhiWDjR2p3p QvELzxRvbLWCOgKu8kdGU1TcGPnkF8eJOPlJVq9PHQg-- X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Fri, 16 Sep 2011 17:07:30 -0500 To: edbark@nist.gov, keri From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "sbvr-rtf@omg.org" , "date-time@omg.org" Ed, A comment about one of your points below. You make some other interesting points that need to be discussed. Ron At 03:34 PM 9/16/2011, Ed Barkmeyer wrote: Ah, but that invites the real question: Must a proposition have exactly one truth value in each possible world? Is it possible that a proposition is both true and false in the same world? Whether a proposition must have exactly one truth value, not zero, not 2, is one aspect of the issue raised. It has nothing to do with whether you know the truth value. That is a pointless diversion. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. That is inaccurate. A person must be in exactly one place _at any given time_. A person can easily be in two places if your world has both yesterday and today. I disagree with you here. I would imagine every natural language has a way to say 'now". In English, you use present tense (as in your example). In Chinese, you would have to say something like "now". To me it makes no sense to take the statement "John is in London" and take it to mean "John is in London yesterday." In conversation, perhaps, but we're not targeting conversational language ... and besides, you have your real-world, one-place rule above. (Interesting that you made a rule ... not sure what to make of that. Rules are simply true (once) by virtue of their being rules.) I believe we must accept that languages recognize a 'now''. If the tool wants to point out that 'now' is likely ambiguous, so much the better. But I see no need to let that issue get tangled up in the meaning of 'proposition'. Ron The example in Issue 16526 provides a fact model in which John is in London is true and John is in London is false (because John is in St. Imier is true) at different times. So, if a proposition must have exactly one truth value in any possible world, 'John is in London' is not a proposition. If a proposition must have at least one truth value in any possible world, then 'John is in London' may be a proposition. Do you want to modify the definition of proposition to say 'at least one truth value'? Was that discussed? Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 4754.78661.bm@omp1019.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316212145; bh=LV58irZ/xihxVz3DED1qssHsRg7MFCe2L0ukMVc2GTM=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type; b=eV2hMQWmSG52LU096KzMO7iQWuO7lYQTEFeQIkYfAScWtIafUrTUeG3YqH6ST0UUzhbrFBA5gtrjWP4NJTAdMMlrpJttnsTQsL5hHYV/nVJlW/fa4QBlqqJMyTVKH+5kScVfhQcG1bA6dKmBP+oSuUofTpRPZFcfWL7rq1fo1zA= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: 9DozmlgVM1n2_4afLo7hl5gMfuZPeyjulYH1LttUz1XVCrz 62NH3.Y1CMQRdHbgofM5hqrGGSxWhikXzcLQTJNNg4M.ixRklj8zFLdUR6iv 84NQZKRnMG1hon19KO.t4XArhwa55Pww8j3q2N2J_rx_YheHo56dYLx4IMcF 48V0et05mXNnQHhc7GbWJmlKIRuKUmQp_gPusy4tHlEXJIhNNEYu5cmdlpdq FUetn57RdXwF5iot61QFj01mmu3kMXqF5RWm3mfwC0OIlxTqDlCtSn1Lxt1y Pm0cFjYOdVH5wQTxT9mRJn609Fby5LL01zWnVxTFkWoP5E66Z8NJkRenh1Yy CgrxCiHLFu2b9xK0k3wF.gpirwZf3Hq6b32sgcgGj9WbDk9BusVZ95jRLJz3 L5087FfJuCh8s6ejf6WRwKR63cU2nj5lPzci4M3qvwNFGUJA8fibHd0PkN1g .ciMXWIQNlq72nxyv9BHVmh8nAZdqMEg_GQxWGCCLdTL4D7s0Vjol6bOxYZB AqcIM3f_rzgSgOyqv6f9CTt5SoItbV1WPllQnE.KvoMAEv62Bk5ITQ21PEtk 6dWc8zdKyP8YpweYapijahKZheMfgMJm738T38Gezf1xWlPig49wOp5FDZHY mgfXPhopx6I_eFmLwXg1S3Xj97wA6XWvvEsyXFWoF X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Fri, 16 Sep 2011 17:28:51 -0500 To: edbark@nist.gov, Stan Hendryx From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "sbvr-rtf@omg.org" , "date-time@omg.org" At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri > wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Date: Fri, 16 Sep 2011 19:12:35 -0400 From: Edward Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: "Ronald G. Ross" CC: "sbvr-rtf@omg.org" , "date-time@omg.org" Subject: Re: meaning of 'proposition' X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edward.barkmeyer@nist.gov Ronald G. Ross wrote: Ed, A comment about one of your points below. You make some other interesting points that need to be discussed. Ron At 03:34 PM 9/16/2011, Ed Barkmeyer wrote: Ah, but that invites the real question: Must a proposition have exactly one truth value in each possible world? Is it possible that a proposition is both true and false in the same world? Whether a proposition must have exactly one truth value, not zero, not 2, is one aspect of the issue raised. It has nothing to do with whether you know the truth value. That is a pointless diversion. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. That is inaccurate. A person must be in exactly one place _at any given time_. A person can easily be in two places if your world has both yesterday and today. I disagree with you here. I would imagine every natural language has a way to say 'now". In English, you use present tense (as in your example). In Chinese, you would have to say something like "now". To me it makes no sense to take the statement "John is in London" and take it to mean "John is in London yesterday." In conversation, perhaps, but we're not targeting conversational language ... and besides, you have your real-world, one-place rule above. (Interesting that you made a rule ... not sure what to make of that. Rules are simply true (once) by virtue of their being rules.) I believe we must accept that languages recognize a 'now''. If the tool wants to point out that 'now' is likely ambiguous, so much the better. But I see no need to let that issue get tangled up in the meaning of 'proposition'. This is a viewpoint thing. When we say: "John is in London." and make the full-stop bold, then it probably means "John is in London now." But in a sentence in which "John is in London" appears as _part_ of the sentence, e.g., "John is in London at least once a month.", there is no assumption of 'now'. As SBVR and Date/time phrase the 'at least once a month' part, they both assume that the 'John is in London' part can be separated, and used to describe states of affairs that play a role in the fact type that relates it to time intervals: 'state of affairs occurs at time'. That is not business English; it is the interpretation we make, because SBVR doesn't support 'adverbial phrases' directly (they would otherwise have to be part of the fact type, as in 'person is in location at times'). When we separate out 'John is in London' in that case, it is still a proposition, I think, but it doesn't mean 'John is in London NOW', because the rest of the sentence (the proposition that uses it) specifies the time intervals. And that is exactly the problem that is at issue. When 'John is in London' is formally extracted to denote a 'state of affairs', is it still a proposition? Does it have to be either true or false? SBVR says it is not a proposition if it isn't either true or false. What if, in two back to back statements like this, it is true in one and false in the other? "John is in London at least once a month." "John is not in London on Fridays." Those are not logically contradictions. That is what the issue is about. The Date/Time team has the idea that no matter how it used, it always describes a situation, whether it exists or not or sometimes. That is, as a meaning, the proposition is the conceptualization of a situation, and true/false is some accidental property of its relationship to the actual world and how it is used. -Ed Ron The example in Issue 16526 provides a fact model in which John is in London is true and John is in London is false (because John is in St. Imier is true) at different times. So, if a proposition must have exactly one truth value in any possible world, 'John is in London' is not a proposition. If a proposition must have at least one truth value in any possible world, then 'John is in London' may be a proposition. Do you want to modify the definition of proposition to say 'at least one truth value'? Was that discussed? [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png *Twitter* || https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." X-Yahoo-Newman-Id: 630993.66239.bm@omp1009.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316218323; bh=v4AkrKP+8iiKe7Df8XLas4koIBpfUwjdlxyohSLG9ik=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type:Content-Transfer-Encoding; b=M1AOzrLydLx30SFZ28cdDHWbUiLVJQkncoD8LHrVJKvkE0VzkkheRKY0Xns6tLHZrZS9ywt2zB3W1JrhV7rb6tg+ndTnoZF/8+3uQImwbiGGuivBHK/NdN+fA6iU6KwsUPcX72krNsQCYY95u+yYizNiNezcM8tI8L55QwaFIlM= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: W0WZKmkVM1l9pYwVxTqPBVXfcjxxma1tGklDljej.7Zl5FQ VOTKeoXrD7DccO.lFwEQXdBkcILeWYYkqEBPrCWVxrw5i2IqAlBTdr_rWGEx bbxRVr9LLHisbYFYKONn0ZJOvBNvpMStzb4ss2oxg43SisrTlf1Xdi.5EAcA 1BiNweYQjiTFWHRgVhbfUwxnIFgz5qYJbfNQBm71HVVvL7fnwPeweekSywZA .Wvmim1YCzskfQsqgucCVGnOAxp41ZeBGZHUmfp5qC.Qw_97VPmBR2XcH.fO dc.ctgKiBRNYADAcL2AiUBvb.z9BW7SQyIWY2LVzv6lPPAcsTi7is5Hiyw8_ TdHM8grUh124Z9CNp6BH6OXTa2ElGxTwK1Qz6IBVSk0AkyAiqgn4KxORgmEu K8_TFXvvPwDxNNgdYy_e09o8bih9c7DrIXCg7tumi4bNsp3R8t4NF.WrXVfT lszS7oUr6vY3ta.JScw.uocDnoxNrB7eqEvwGY5RO6X3lMxitMPNFp6BhKHT VzjGAmc89RjG1lb8wAhH1qLLOgKiNqkXvjdoT18GC9kWRppLHQsVR43rKQZu RbTCAVbXn0lZv_TEKtE42HBFwjoZzw3JbminpLScV9EN0gZwc1DUEHXuI291 gXS6LmvkdodQuL7vVa6FKPInko1NYFb9opcNrNZasJyrM7XNuvX4M9Eo_LlF 7YEzLLFEcPxoVvxU7yQ-- X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Fri, 16 Sep 2011 19:11:50 -0500 To: edbark@nist.gov From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "sbvr-rtf@omg.org" , "date-time@omg.org" At 06:12 PM 9/16/2011, Edward Barkmeyer wrote: Ronald G. Ross wrote: Ed, A comment about one of your points below. You make some other interesting points that need to be discussed. Ron At 03:34 PM 9/16/2011, Ed Barkmeyer wrote: Ah, but that invites the real question: Must a proposition have exactly one truth value in each possible world? Is it possible that a proposition is both true and false in the same world? Whether a proposition must have exactly one truth value, not zero, not 2, is one aspect of the issue raised. It has nothing to do with whether you know the truth value. That is a pointless diversion. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. That is inaccurate. A person must be in exactly one place _at any given time_. A person can easily be in two places if your world has both yesterday and today. I disagree with you here. I would imagine every natural language has a way to say 'now". In English, you use present tense (as in your example). In Chinese, you would have to say something like "now". To me it makes no sense to take the statement "John is in London" and take it to mean "John is in London yesterday." In conversation, perhaps, but we're not targeting conversational language ... and besides, you have your real-world, one-place rule above. (Interesting that you made a rule ... not sure what to make of that. Rules are simply true (once) by virtue of their being rules.) I believe we must accept that languages recognize a 'now''. If the tool wants to point out that 'now' is likely ambiguous, so much the better. But I see no need to let that issue get tangled up in the meaning of 'proposition'. This is a viewpoint thing. When we say: "John is in London." and make the full-stop bold, then it probably means "John is in London now." But in a sentence in which "John is in London" appears as _part_ of the sentence, e.g., "John is in London at least once a month.", there is no assumption of 'now'. As SBVR and Date/time phrase the 'at least once a month' part, they both assume that the 'John is in London' part can be separated, and used to describe states of affairs that play a role in the fact type that relates it to time intervals: 'state of affairs occurs at time'. That is not business English; it is the interpretation we make, because SBVR doesn't support 'adverbial phrases' directly (they would otherwise have to be part of the fact type, as in 'person is in location at times'). When we separate out 'John is in London' in that case, it is still a proposition, I think, but it doesn't mean 'John is in London NOW', because the rest of the sentence (the proposition that uses it) specifies the time intervals. And that is exactly the problem that is at issue. When 'John is in London' is formally extracted to denote a 'state of affairs', is it still a proposition? Does it have to be either true or false? SBVR says it is not a proposition if it isn't either true or false. What if, in two back to back statements like this, it is true in one and false in the other? "John is in London at least once a month." "John is not in London on Fridays." Those are not logically contradictions. That is what the issue is about. The Date/Time team has the idea that no matter how it used, it always describes a situation, whether it exists or not or sometimes. That is, as a meaning, the proposition is the conceptualization of a situation, and true/false is some accidental property of its relationship to the actual world and how it is used. Thanks, Ed. Those are great examples. I understand much better now. Let me venture this (and quickly duck). Seems like the error here is over-reduction. In other words, a time-specific proposition cannot be reduced to a non-time-specific proposition. (I purposely say time-specific, because "now", "always" and "never" do not apply.) A time-specific state-of-affairs cannot be indicated to include a non-time-specific state-of affairs. The proposition "John is in London." cannot be properly reduced from the proposition "John is in London at least once a month." If don't know whether the proposition "John exists." can be properly reduced from "John is in London at least once a month." However, I'm sure the proposition "John is in London at least once a year." can be. If I'm off-course, I'm sure someone will set me straight. This is just one set of examples. I do see there are some issues with the definition of proposition. However, I think a 1-to-1 alignment of propositions and states of affairs is very important. Seems to me all bets are off if that isn't maintained. I can imagine how it would work it they aren't essentially mirror images. Ron -Ed Ron The example in Issue 16526 provides a fact model in which John is in London is true and John is in London is false (because John is in St. Imier is true) at different times. So, if a proposition must have exactly one truth value in any possible world, 'John is in London' is not a proposition. If a proposition must have at least one truth value in any possible world, then 'John is in London' may be a proposition. Do you want to modify the definition of proposition to say 'at least one truth value'? Was that discussed? [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || http://www.RonRoss.info/blog/ < http://www.ronross.info/blog/> http://www.brsolutions.com/email/linkedin.png *LinkedIn* || http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || http://www.RonRoss.info < http://www.ronross.info/> -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 724813.57354.bm@omp1019.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316219713; bh=M9QFEj5Aauu0oDxoSQnAlbImwC20qTpsGgOcU6kMb8k=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:References:In-Reply-To:Mime-Version:Content-Type:Message-Id:Content-Transfer-Encoding:Cc:X-Mailer:From:Subject:Date:To; b=YnWiD35831Dk850QnlE23FUPK2JPilM0lON//gnaqHTzkJr5xO24cYzBjeuOpXb2vhM3ysKYnOWESBdL9s83jfB1xAYcwMIFM0ubPw4n17df/Su1y7ChLfYHoCvozdUTxccVP9XhZEwSeCywICVDcZJLgA2GA+F0dvhAk5bJ57Y= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: BEeKXYEVM1kvH90EUyz04z1AleUEngvwekCPUNgqkZ4PxmD 3Z3XnGuXkWsWobLjyJqw7kGC_1nc2F1qKgEZoqIcXIY5DU8jvvsudoOeuycZ sCIuauZM5lb99fPIv7LcUcDcdLTYjWabFHiBne3xr2p1HTSEblrROx80CfBD mLHjBtVee1.LTkTF40cfFBI.sHyDy0vI4mVoLZA5EdJRhxpzv_UU3egYxI_T 8d49NHE_JDvj4S9LDOWrMPYGBDrnKetyYksfvjjcM2.R5ik8PHZBh3T5Qe8l jTqtBOK1K1vg8tUYa.AnWuSXXtLpJ9Aaf8YeEADyVTv0wNI3zjnKZtMUDf8. EsH0xuPhIcVZUKVKCf18bjG9v7izEd.cdNErxQHZIgfNa92cy1WPO4jI.ngp uxSIeP0ZuoBt9PeFngvoTJpa8AHObLOK8ara6DvK_Jid79VvRlF1zZLFGKXb MX8Z1nudD.OC4kGx_vv8DhhtQNugReeHUY84UoW7UAcC2Y4nR_hhPEbYj0__ QFX01EXtnB932QlWx7REWhNuw16XGctIhNUZNF4iWmpNluvh_goJU8YRfxH4 Rn6b8g7DgizvD34lBlb_ldO9wQhFcorK8m3ZNsSR.vHIjQZ69u0k673ingQu jyKegFn5.P1_l.uUG6g-- X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI Cc: "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" X-Mailer: iPad Mail (8J2) From: Stan Hendryx Subject: Re: meaning of 'proposition' Date: Fri, 16 Sep 2011 17:38:34 -0700 To: "Ronald G. Ross" Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri > wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Date: Fri, 16 Sep 2011 21:20:18 -0400 From: Edward Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: Stan Hendryx CC: "Ronald G. Ross" , "sbvr-rtf@omg.org" , "date-time@omg.org" Subject: Re: meaning of 'proposition' X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edward.barkmeyer@nist.gov Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png *Twitter* || https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." X-Yahoo-Newman-Id: 76224.62117.bm@omp1016.access.mail.sp2.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316240161; bh=Qk0pukADl/T4uv1PH37ItvY304U3flou8MPnVev+SKQ=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:References:In-Reply-To:Mime-Version:Content-Type:Message-Id:Content-Transfer-Encoding:Cc:X-Mailer:From:Subject:Date:To; b=EdjlyoCF3UgZYKojvCZVP/zd+UTWwMTFsdj0okcLXP8vddkKL7zmAEIATlzux5fGoHWxvtMEJnqKbOKlk4mLpfrUBZlKLPFF/7TyitbzP80Q+HqV/ghKs1/tf6e9C2sF9e4YXFNLrlQ5bcNPtImWPL7DW0xtJJ8ZQiTb8moG33E= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: 3IDsz8sVM1kE48dLCxGVSeW7waef0YxqgsAJaIS2tPRNSTr E0C9x51saq_3UAtOieaAc6.BBzmGKbyCSdogP3jYxqyugyQX864OpD9umPDW NXn1335H6dUGonf_51njw.hx2lQGqxqdxuF8JccY3Kg_fYgJoqsNP8bhCMXk 2Hrffb0x3HCZGWups6MwrqOsiIcgQ4veHelz1FzUH9ZNNOufmDuzD.9c2zOm 8.H94JRwHkE_jo_cJ9.ZNFVNmBRup70uIY3AqSWemkhoF4F2fG68gEFHCdRH M4r4LiqxvELLgYQT_VhjmSKMcYa6nQPUt1uKm4q.sw1Vsfw4NFLbh.lRsW30 7XaXMbMJNkRzKSitfaGQFjEBULFzqhf_b93RbCv4FW6XR7P0a7PNtaD0GKZa kVpkj1tT9aj61dYMUCYbaO50K9Ul1hRDtM1KLD1IcxA3jWDZXyoghfqmX8h0 jMnkPXGKV23qLW8yBKk2eCDVr4JSoHexuq2JOpQFr.R9fI.n7038R0pOR1zm 4Q4Y5Oq3hYOfc9Yfvui0iuC7p8xtmyrnhHOJrMc2xzTBaW1Eh67X6zW6bblw 7jn9l7fWvan85cDzjx_ZQSOl3N2HYzaYLSSoYpica9zfSUho85ATP2kfWR4D 9ZcFAZVDiG1sPY0xJmQ-- X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI Cc: "Ronald G. Ross" , "sbvr-rtf@omg.org" , "date-time@omg.org" X-Mailer: iPad Mail (8J2) From: Stan Hendryx Subject: Re: meaning of 'proposition' Date: Fri, 16 Sep 2011 23:19:23 -0700 To: "edbark@nist.gov" X-MIME-Autoconverted: from quoted-printable to 8bit by amethyst.omg.org id p8H6GE7q031690 I think Ed and I are in agreement that the SBVR metamodel can formalize statements in many different natural and formal languages. Thanks for pointing that out, Ed. We also agree that the formal semantics of Clause 9 could use some work. Its potential for multi-lingual formalization has always an attractive feature of SBVR. ODM is a significant contribution in this direction. Largely because it admits modal logic, Henkin higher-order logic, and soon Date-Time, SBVR is more expressive than most other popular formal languages, including CLIF and OWL. Stan On Sep 16, 2011, at 6:20 PM, Edward Barkmeyer wrote: > Stan Hendryx wrote: >> Ed wrote: >>> Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, >> >> Three SBVR fact types are particularly relevant here: >> statement expresses proposition >> closed logical formulation formalizes statement >> closed logical formulation means proposition > > Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. > >> These are related by this rule, which is a composite of several SBVR necessities: >> Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. > > A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. > >> This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. >> >> A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. > > Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. > > A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. > > This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. > >> Ron wrote: >>> "Truth conditions" is new to me in this context. >> >> This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >> >>> Certainly logical formulations should be considered out of bounds for defining what a proposition is. >> >> On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. > > I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. > And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. > >> To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. > > I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). > >> I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? >> >> If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, > > I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. > The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. > > -Ed > >> since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. >> Stan >> >> On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: >> >>> At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: >>> >>> >>>> Stan Hendryx wrote: >>>>> Keri wrote: >>>>>> What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. >>>>> >>>>> Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. >>>> >>>> Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). >>>>> The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. >>>>> I hope this helps clarify what Ron meant. >>>> >>>> Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. >>> >>> Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. >>> >>>> We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? >>> >>> "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. >>> >>> I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. >>> >>> Ron >>> >>> >>>> -Ed >>>> >>>>> >>>>> Stan >>>>> >>>>> >>>>> >>>>> On Sep 16, 2011, at 10:59 AM, keri < mailto:keri_ah@mac.com>> wrote: >>>>> >>>>>> Mark, >>>>>> >>>>>> Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) >>>>>> >>>>>> What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. >>>>>> >>>>>> I also captured points such as: >>>>>> * Being true or false does not depend on (a) if it's known that >>>>>> it's true or false or (b) that it's interesting that it's true >>>>>> or false. >>>>>> * Being true or false is different from ambiguity. >>>>>> >>>>>> In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. >>>>>> >>>>>> - Keri >>>>>> >>>>>> >>>>>> On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: >>>>>> >>>>>>> During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. >>>>>>> >>>>>>> > Given a fact model comprising: John is in London during 5-8 September >>>>>>> > 2011. John is in St. Imier during 9-10 September 2011. The time of >>>>>>> > interest in the Universe of Discourse is Sept 5-10, 2011. The >>>>>>> > proposition expressed by 'John is in London' is used in representing >>>>>>> > this fact model, both in SBVR examples, and formally in Date/Time. Is >>>>>>> > it true or false? I believe that in that actual world, the proposition >>>>>>> > is neither true nor false. Therefore, it is a counterexample to what >>>>>>> > Don says. >>>>>>> >>>>>>> 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. >>>>>>> >>>>>>> 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. >>>>>>> >>>>>>> >>>>>>> Is this a correct summary? >>>>>>> -------------------------------- >>>>>>> Mark H. Linehan >>>>>>> STSM, Model Driven Business Transformation >>>>>>> IBM Research >>>> >>>> -- >>>> Edward J. Barkmeyer Email: edbark@nist.gov >>>> National Institute of Standards & Technology >>>> Manufacturing Systems Integration Division >>>> 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 >>>> Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 >>>> >>>> "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." >>>> >>> >>> [Ronald G. Ross Contact Information] >>> http://www.brsolutions.com/email/wordpress-2.png *Blog* || http://www.RonRoss.info/blog/ >>> http://www.brsolutions.com/email/linkedin.png *LinkedIn* || http://www.linkedin.com/pub/ronald-ross/1/3b/346 >>> http://www.brsolutions.com/email/twitter.png *Twitter* || https://twitter.com/Ronald_G_Ross >>> http://www.brsolutions.com/email/button-green.png *Homepage* || http://www.RonRoss.info >>> > > > -- > Edward J. Barkmeyer Email: edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 > > "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." > X-Yahoo-Newman-Id: 416207.60890.bm@omp1023.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316298155; bh=qEYnPW5xT4KBoGME1BozCMcjLStJiAFmXQRYehSUY1Y=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type; b=0AirDROdQEZyytJSLUGHYDKqQIMW+IJ1G/caDi56+tXPajj7PHmV5vQGIKATtNEespDba6DLU1rGEuN0NIdMPKaPs+fOpAQBLNtR1zPHo++cBoSN1kuOF8aQIqS2jn6/UfZCB+tQ7XWH+G2RjdQiRVnwcdWjvA7R0BZJdHgCd9g= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: aV2Yx7gVM1m_aVcYYsMJk6VCLEptX4BFT2gKejSJU0KhqkT _nNpwojZbT2wjNWuMj9xtKeMMuWYtgIzuSJeGwRq9NrWSnLIUgKlc4pqNE9c XnhjSfOsR8xU3z.Rc43zb7eYzxTAa7pz13fnLY5mv7gGtX_WE670o_ZrkDU8 waCB73a_eNoQe49bPe21PPLRt52Q4KAtSGCXW4EhM6FP2Mx44koVzLSnMUUf zVDsBMzwN6SUluTLfYRr8YkizTYhHEUyT_a1ZL_mSPTTXALHXZgqkz023AjT _GlmZZV959VaWjiY2Pd0IKth8hT5NkHof5K6aG0vvRNQpxNd15ihdCjx.llR NdXYMdmVMmYKC9bLL3Or_r4fN8CkcCzf6Lh1OpgjwwAoyNdOGFduSg1WMbZc r5hiyvh0koDVqTIiOhlYIedIf9v3Z9B_R3MlLzyb7QTPmmsqxf8tvkHz26Pt W1PnKm8wiZXMG.D3Z4KFrdYsfIZflvxgLAKRTX3RLJKidi7ZKARyaJbHk9A2 coB1XFQYNrplf_3891I94gWQmbefRpvpGdhrsyQRKVFzp_10.3NlTthm0a2p LRodxIP.PwYZcxfEeuZSaBzknHK476.kvWsUm7uf1AhhU3XnEGyEcq2xaVFF _VBfnmGm5Y4ACR99Iu4R47ZJ4wZs9E9VRP0NyRTvVS1Qq3Rw1Xap63DECJHy cmSAccod3bVN2NSt4 X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Sat, 17 Sep 2011 17:22:24 -0500 To: edbark@nist.gov, Stan Hendryx From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "sbvr-rtf@omg.org" , "date-time@omg.org" At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. Ed, I'm not sure why that should be an annoyance. SBVR was always about business vocabularies, business definitions, business rules, and business languages ... what businesses need to communicate with each other and with machines in structured natural languages. That's its purpose and its scope. Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Wouldn't that be reinventing the wheel?? Sorry, I don't follow ... but it would seem to be a very important point. Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 991499.82954.bm@omp1010.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316299521; bh=Wl0V5DxtteamEyFOY6zP1AF6QywkL6qzo9h2vS52cMI=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type:Content-Transfer-Encoding; b=LDx3H60AV0BMMu/fUBJT9qKcPBb3VBRRreipxJJFdhSCLzUvefRrPIOX3LaT+lGwv+bCbC5yPBH5/jvaU2YpFWl/o+IlPtiiSMWNfTvlCfBxqL8TGeYJyyJpmLWlykJc6MJqTgfynp+9sYydhy22a/oby9k2An9ZqWdY4cXtZVk= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: lxoEwZIVM1lFB8Bu4edwebYx7mh4F7.hP0C02SdqcdMey8j T.VIidDuzqttgCln_m1zzlays3RKEQcLG2QI7w6rcwRBicXJq0QyN93NdXBZ eQjjjyfRcpjYhrxoy0fEFRAe0WtsAt.xWeZdyZxKct136Zjrf3URbiksKveO 1h7dCMFgezSw32MHE8QDwiLFTuhGSZVpOEjGYdMrbrBMdTU8h5l6YkowFq3T 3sML8gqcCAcvh44dS2vBQpcrTixWeG9Tmd0EEmA.hVvgPyGo0MjDsWVwo5KS raHjYt8Hpn7rnXFbUvTTPehJSNp20D3UwiFTVYCGfAhFqYQa6QKZW73WSEfr BxqoZgOb_pwu8nC.ArXCUWdz5t6dEdOm1L7UHzngka6WsLl1liHo3wBJX4yz 8a7tlo.c_m9HLlFYHEBGagr1ZyXLlvXLo3oleRQFbqh4PqWvfKAkYYMnnIvB b1Yoox.eYybh8lpqEOV6kNfwhRv99l.kFGojXMHW8Qfp_LPdb2KchV8e12TB NSYNWRL.VdW6arRqMhwBAtZH.AwtwRjQuy16IQ6oTM9GwT83HE.MGxTBpy8C rgRNaVEEgo2hqr_J8acnUXr55egS7n8E- X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Sat, 17 Sep 2011 17:45:10 -0500 To: edbark@nist.gov, Stan Hendryx From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "sbvr-rtf@omg.org" , "date-time@omg.org" >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 248876.93305.bm@omp1025.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316331948; bh=+nd2lCj0Wz3N9PPQsoGj8Z/o/iQZ+thi0tJVXrPoVMg=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:References:In-Reply-To:Mime-Version:Content-Type:Message-Id:Content-Transfer-Encoding:Cc:X-Mailer:From:Subject:Date:To; b=H0XW+gClxl8x5HDwy1LzCfBJxCBOXKyAiOun/TcNzCWGqPj6PxXw9nVRObzN699nCJKIgOupOvQa0IMYozrF+n995E0NrEXOdTh2gOuQ5PPi9l+BpI6QUI6Zyfa3XFW39bMOomg83kzb4G2uV8TH1Pekzh7z2x+LRjTaw5lapww= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: 8P2jwnQVM1lvG.hiJCPfPbpthyOkEmpMw8yLrm1_kSzuXnv yfET51jUdVEImGiAJY8001om3fsGYKyMKV6pxm4heLiMYi._Czq6CkSYtleZ 5MCAPHne.0Yvm2.df0ye3idXlKzq6nEKCqTH4gPVk5ikeZoSYCBn42CG.LY9 e1952gaGNjT4OGCpS6D.ADF9_CdmB2slKgDdd.NtnR6UpNVIoXV6b89w3f14 Aq8NjgKwOLwgV1a4sL2nN.Y5n8bwnA9PNX.IVVgYx.V0LHgbDkvHeD37ricA SXPzKoGjlNW4zA1JrCOJYznNs6pT8IYJu3k8_hicSFhDCn6IcptU_hesd46n P4X_hMPLvrqT_4zNwWfeEFNfALYAGCgLh4tqB3SY9NOSudUm.Dj84ieQckXv g40NE80hNKalw3renebNMWNdYeAmWYqXSYwkwuLq0mzCcFLOdsr3fe.qgFbN URJvgE9k0O.JCH72nqD3MvSFahXFhyVoAMhkXXd9zhVpXh0eWVbTorMAjQ8D ixQnesBiV81lY35A3FyjgHhzykh_E55rve0H4aKELgVNXVyU7sgKCdsaS0Ew zGZNY3m.yXvMAyFA0UavhV8f5Evs8.UYqA1g5gx7cHGCYLQaGcPFPkuoBWiM - X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI Cc: "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" X-Mailer: iPad Mail (8J2) From: Stan Hendryx Subject: Re: meaning of 'proposition' Date: Sun, 18 Sep 2011 00:49:11 -0700 To: "Ronald G. Ross" Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. >>What about existential propositions? e.g.., John Hall exists. These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), pp. 25-50 Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. If a fact model meets the truth conditions, the proposition is true for that fact model, for the possible world modeled by the fact model. What it means for a fact model to meet the conditions of a proposition is spelled out in the model-theoretic semantics of SBVR, which, as Ed pointed out, is not well documented. This is an area for future work on SBVR. The model theory of a language is a significant part of it's formal grounding, important to interpreting the language and getting translations correct. Not all 'artificial' languages have a model theory, e.g. UML does not. Stan On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" wrote: >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 706002.10096.bm@omp1004.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316356874; bh=cQ28+NwN88KBy0cwqUGfCkYjWzdpmjSj6DgBcmy9UD0=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type; b=eSsERpsbRzCdyfWgHKUXe4qLaiw3oKduIpLgRyzQkZ0Vvf9DFu4YY0uTltTt9LVwCnzY4bgOqNsBEGa/nOV1UuvKIwGxEog9/A/3nb+HMnfCjFul9guv6Qe5q2+aeowtNdPYT0HGnYBMNpunUiWC+59pH/Te4rnpK+hCS4v5Eus= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: tTAEfbEVM1kLHttiCnCPS6Xj7FZfU7Srj_n.2KeJxsB.4k_ SeomfYAfu81eSoZVE7xrohNlwbv1Qb2XVieCISx5u5lb_RZkLqktjW1M2KZL cMNYjF1AZG9QMHda8eQVJzl.UYZ7hfzk8w4dJQUgPevtLuGS02m.IAyMIFCz Z_jFKHSmqUHPQfB30RdPT7X4YB7fQSzCncogb551FCg0wtwqL7eYPDp_fdg4 RGLSzGl7_pBZkoyxuDD0RBq9T_WGeZJNfJxLAhGUj6QmJM.DqkWfSBud2dEv XCfkWTDcuXRpPWalWzEgBNoLBwTFOWR4.3b1vijt0dmE7Oj8JnvZRIEyyuyz GDMUle.uCFNx69WcyrIyoOP1A294BRuUnDieicyaKGIQVH3xDXul9v7aMI.y iJUr9.TRrWCfHvTBqxAC1Q8yALn_1F3Xhnurjj7m7JSZjcTQVmQAqK2AwKB6 SdzObTysECDagXOIg995QJP9WejfOi64eFoiQ8_YmO.N3PfABVcsBMphGlF1 T43WPkakhAJqzMhCOJ2vpfQYkUF6urKyl.tFZjNOlkxXeXuWAF9vrVD_c3_Y yxPwYaySYjYwi9UO2n7o6bS8bQ6vBBf2XIm75I17TELBL5k3QulVsWoVkKdL 5VFeSHOtF01_1hWzF9UEBuw.g_Afl7bHIWwIazaVSGMt82bZcQZG4Vbrz.U. e42B5OJX6JpEjgXFC X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Sun, 18 Sep 2011 09:41:09 -0500 To: Stan Hendryx From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" At 02:49 AM 9/18/2011, Stan Hendryx wrote: Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. To me that's outside the scope of SBVR, but I defer the question to others. Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. To clarify, all I meant was I (and I mean only *me*) hadn't seen the term in recent discussions with the d-t team. Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 706002.10096.bm@omp1004.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316356874; bh=cQ28+NwN88KBy0cwqUGfCkYjWzdpmjSj6DgBcmy9UD0=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type; b=eSsERpsbRzCdyfWgHKUXe4qLaiw3oKduIpLgRyzQkZ0Vvf9DFu4YY0uTltTt9LVwCnzY4bgOqNsBEGa/nOV1UuvKIwGxEog9/A/3nb+HMnfCjFul9guv6Qe5q2+aeowtNdPYT0HGnYBMNpunUiWC+59pH/Te4rnpK+hCS4v5Eus= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: tTAEfbEVM1kLHttiCnCPS6Xj7FZfU7Srj_n.2KeJxsB.4k_ SeomfYAfu81eSoZVE7xrohNlwbv1Qb2XVieCISx5u5lb_RZkLqktjW1M2KZL cMNYjF1AZG9QMHda8eQVJzl.UYZ7hfzk8w4dJQUgPevtLuGS02m.IAyMIFCz Z_jFKHSmqUHPQfB30RdPT7X4YB7fQSzCncogb551FCg0wtwqL7eYPDp_fdg4 RGLSzGl7_pBZkoyxuDD0RBq9T_WGeZJNfJxLAhGUj6QmJM.DqkWfSBud2dEv XCfkWTDcuXRpPWalWzEgBNoLBwTFOWR4.3b1vijt0dmE7Oj8JnvZRIEyyuyz GDMUle.uCFNx69WcyrIyoOP1A294BRuUnDieicyaKGIQVH3xDXul9v7aMI.y iJUr9.TRrWCfHvTBqxAC1Q8yALn_1F3Xhnurjj7m7JSZjcTQVmQAqK2AwKB6 SdzObTysECDagXOIg995QJP9WejfOi64eFoiQ8_YmO.N3PfABVcsBMphGlF1 T43WPkakhAJqzMhCOJ2vpfQYkUF6urKyl.tFZjNOlkxXeXuWAF9vrVD_c3_Y yxPwYaySYjYwi9UO2n7o6bS8bQ6vBBf2XIm75I17TELBL5k3QulVsWoVkKdL 5VFeSHOtF01_1hWzF9UEBuw.g_Afl7bHIWwIazaVSGMt82bZcQZG4Vbrz.U. e42B5OJX6JpEjgXFC X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Sun, 18 Sep 2011 09:41:09 -0500 To: Stan Hendryx From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" At 02:49 AM 9/18/2011, Stan Hendryx wrote: Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. To me that's outside the scope of SBVR, but I defer the question to others. Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. To clarify, all I meant was I (and I mean only *me*) hadn't seen the term in recent discussions with the d-t team. Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Date: Mon, 19 Sep 2011 08:00:36 +1000 Subject: Re: meaning of 'proposition' From: Terry Halpin To: Stan Hendryx Cc: "Ronald G. Ross" , "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" <> I disagree with Stan here. I use "role" in two ways. A "fact role" is a part played in a fact, and always involves a logical predicate (e.g. The fact type "Person drives Car" has two roles, one being the role of driving which is played by a person, and the other being the role of being driven which is played by a car). A "role type" is a non-rigid type (e.g. Student), unlike rigid types such as Person or Thing. The elementary fact "John Hall is alive" involves an individual and a unary predicate "is alive", which may be treated as the role of being alive. The existential fact "John Hall exists" involves an individual and an existential quantifier. The usual view in logic, with which I concur, is that existence is not a predicate (and certainly not a first-order predicate). "Exists" is not the same kind of construct as "is alive", and it does not involve a role. If we use the individual constant "j" for John Hall, then "John Hall is alive" may be formalized as "j isAlive" (in postfix notation), but "John Hall exists" is formalized as "Exists x x = j". Cheers Terry On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx wrote: Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. >>What about existential propositions? e.g.., John Hall exists. These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), pp. 25-50 Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. If a fact model meets the truth conditions, the proposition is true for that fact model, for the possible world modeled by the fact model. What it means for a fact model to meet the conditions of a proposition is spelled out in the model-theoretic semantics of SBVR, which, as Ed pointed out, is not well documented. This is an area for future work on SBVR. The model theory of a language is a significant part of it's formal grounding, important to interpreting the language and getting translations correct. Not all 'artificial' languages have a model theory, e.g. UML does not. Stan On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" wrote: >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 886400.85427.bm@omp1014.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316388803; bh=WhF3cBbiw8SnI8cCCjXvUobiqETSwSpvuu3F5JL5HzM=; h=Message-ID:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:X-Mailer:Date:To:From:Subject:Cc:In-Reply-To:References:Mime-Version:Content-Type:Content-Transfer-Encoding; b=2+9ZjKrfLAqMqiH+lKXQDElLS6sw+T2a6hD+gzI/aM1Fgo4FEonCdx6/AV10I3x/aND0G7WPwgUx7sgWO8djLMHxSDo/dca4k5abmygjSlRs49NOwZMcCHEi1+onr0V0jPm3nI7RXZsCP7lrZaCR0letcvV7nFPkSIqmBGYj0qw= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: Du1PyUgVM1nD1mRgeHOOssGjjFfdQLhcZ9C3w.1ObXmB.vA LkwGFOPXpTJ4.4f7QFrpD7DAQFrymvEe.GCSAAmTqFLdBE_2afq5gE_xcNa5 4tyIuG0e5uSF2N9U26kKhxtQPdzyrGUNk2Oxo2tzF8Lp17Se_AAkOsOANPXR aDid7_KTtG4GTmySuVxKcqtUk8dqvgDVWJGFE3tUze8ba2KAony2J0ZadbV2 O4SrM11U8ddMh0_gtgYBCoPhNPWyFHscQC._IPonSD9Oki6hsDhGfEXMgd1u OWfWv5chgX.AMatlSSscaYENSBWB9qRqnAHPPQwN_4aLgKelXSihwaxPYPlH jtAsXprJKHEhnfiszxWEyhIcPOZpo52WRpm0HAS8L1m4gCA-- X-Yahoo-SMTP: MhfrpU2swBDLgYiYhNQDHBu0cE4o.vu2We1FRN9o X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Sun, 18 Sep 2011 18:33:02 -0500 To: Terry Halpin , Stan Hendryx From: "Ronald G. Ross" Subject: Re: meaning of 'proposition' Cc: "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" At 05:00 PM 9/18/2011, Terry Halpin wrote: <> I disagree with Stan here. I use "role" in two ways. A "fact role" is a part played in a fact, and always involves a logical predicate (e.g. The fact type "Person drives Car" has two roles, one being the role of driving which is played by a person, and the other being the role of being driven which is played by a car). A "role type" is a non-rigid type (e.g. Student), unlike rigid types such as Person or Thing. The elementary fact "John Hall is alive" involves an individual and a unary predicate "is alive", which may be treated as the role of being alive. The existential fact "John Hall exists" involves an individual and an existential quantifier. The usual view in logic, with which I concur, is that existence is not a predicate (and certainly not a first-order predicate). "Exists" is not the same kind of construct as "is alive", and it does not involve a role. If we use the individual constant "j" for John Hall, then "John Hall is alive" may be formalized as "j isAlive" (in postfix notation), but "John Hall exists" is formalized as "Exists x x = j". Terry, Would it be fair to say then that there are two categories of proposition (relevant to SBVR) -- (1) existential propositions, and (2) propositions involving some role? The current definition of "proposition:" in SBVR seems inadequate ... "meaning that is true or false " ... and this is one source of much consternation. For example, that definition would seem to admit boolean variables. I think we really need something better. Thanks, Ron Cheers Terry On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx wrote: Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. >>What about existential propositions? e.g.., John Hall exists. These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), pp. 25-50 Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. If a fact model meets the truth conditions, the proposition is true for that fact model, for the possible world modeled by the fact model. What it means for a fact model to meet the conditions of a proposition is spelled out in the model-theoretic semantics of SBVR, which, as Ed pointed out, is not well documented. This is an area for future work on SBVR. The model theory of a language is a significant part of it's formal grounding, important to interpreting the language and getting translations correct. Not all 'artificial' languages have a model theory, e.g. UML does not. Stan On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" wrote: >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Date: Mon, 19 Sep 2011 09:54:19 +1000 Subject: Re: meaning of 'proposition' From: Terry Halpin To: "Ronald G. Ross" Cc: Stan Hendryx , "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" Hi Ron <> I distinguish purely existential propositions (e.g. there is a country code 'US') from propositions that predicate over individuals (e.g. Ron was born in the USA). To me, a proposition is what it is that is asserted by a declarative sentence (or the declarative component of a sentence), and is always true or false (but not both). The SBVR "meaning that is true or false" seems to correspond pretty close to this. I don't consider a variable to be a meaning. Cheers Terry On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross wrote: At 05:00 PM 9/18/2011, Terry Halpin wrote: <> I disagree with Stan here. I use "role" in two ways. A "fact role" is a part played in a fact, and always involves a logical predicate (e.g. The fact type "Person drives Car" has two roles, one being the role of driving which is played by a person, and the other being the role of being driven which is played by a car). A "role type" is a non-rigid type (e.g. Student), unlike rigid types such as Person or Thing. The elementary fact "John Hall is alive" involves an individual and a unary predicate "is alive", which may be treated as the role of being alive. The existential fact "John Hall exists" involves an individual and an existential quantifier. The usual view in logic, with which I concur, is that existence is not a predicate (and certainly not a first-order predicate). "Exists" is not the same kind of construct as "is alive", and it does not involve a role. If we use the individual constant "j" for John Hall, then "John Hall is alive" may be formalized as "j isAlive" (in postfix notation), but "John Hall exists" is formalized as "Exists x x = j". Terry, Would it be fair to say then that there are two categories of proposition (relevant to SBVR) -- (1) existential propositions, and (2) propositions involving some role? The current definition of "proposition:" in SBVR seems inadequate ... "meaning that is true or false " ... and this is one source of much consternation. For example, that definition would seem to admit boolean variables. I think we really need something better. Thanks, Ron Cheers Terry On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx wrote: Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. >>What about existential propositions? e.g.., John Hall exists. These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), pp. 25-50 Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. If a fact model meets the truth conditions, the proposition is true for that fact model, for the possible world modeled by the fact model. What it means for a fact model to meet the conditions of a proposition is spelled out in the model-theoretic semantics of SBVR, which, as Ed pointed out, is not well documented. This is an area for future work on SBVR. The model theory of a language is a significant part of it's formal grounding, important to interpreting the language and getting translations correct. Not all 'artificial' languages have a model theory, e.g. UML does not. Stan On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" wrote: >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 427232.15243.bm@omp1014.access.mail.sp2.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316421703; bh=8uNsYjUTM0ymK6e/iThrIYFLb9qz/Yk+wdYXGwIw1ig=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:References:In-Reply-To:Mime-Version:Content-Type:Message-Id:Content-Transfer-Encoding:Cc:X-Mailer:From:Subject:Date:To; b=UfXyWHNxtpHsOaor1DrQtFSDlANGYbQ67HIa4uQkR14WvTUEVcmUkXrXsUh9y1mlVEtjCR8Uvzy3IVsMH9PrP9/6QZlHlO+UUqi/veh8mG0IOaILTIqjIYEZlqfOkM9yZra16fE0eCR1tW0Gk8xbw/0JHd4H3QZGIohT+x+mdiM= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: ZVkSMeMVM1mPoIQZW4bJ.yf2aH1BpCCUS9YE0jSZoTdp.y3 SiCmBnWhQ_4YgsruXhup9NsXfrWiu_RXiIs7L10.OZ2L3NPSNNmPhQj4qHm4 RiRTL8Ys8dIOgvKIPPAZYMsIUJoJ1EVKgnHzDqnmbFAAzDGNsMJXyH_LOMem ZuDOCUmHJ4LxWYKa5lrUg0ECt5V4boMMOu_ENWxmbzmEk4nxaiQH6Wppkn_D hK3mBDC76IaqLuGx3hZ8MbIq9BgpW6babZZ4Qh.n5UtV9aJihTJH4augtmPB mNTpf.rAz_FXqm5yTI9.Oiu53XOtIrDyUuXvWjXE9CWI0bXMztl25cnIkrZ. 7eHFGPF51_7RLmiGfjJxA7jJUfPC3EazQ3ZsIfFb.D70u8WvM7RdgMTy8ezM p0_XCi01W0QGZoImqkRMgV.8N_Q-- X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI Cc: "Ronald G. Ross" , "edbark@nist.gov" , "sbvr-rtf@omg.org" , "date-time@omg.org" X-Mailer: iPad Mail (8J2) From: Stan Hendryx Subject: Re: meaning of 'proposition' Date: Mon, 19 Sep 2011 01:45:07 -0700 To: Terry Halpin Terry wrote: "John Hall exists" is formalized as "Exists x x = j". Yes, I agree. I am corrected. "Exists" is not a predicate. "j exists" is an elliptical form of "Exists x x = j". In SBVR, "j exists" is formulated as an instantiation formulation, which is a logical formulation that considers a concept (person) and binds to a bindable target x with the meaning: there is x such that x is a person and x is j. "Exists" is based on the fact type 'thing is thing', which has two fact type roles. Thus, in SBVR, existential propositions involve roles. In SBVR, all propositions involve roles. In general, an instance j of a concept c is a state of affairs that corresponds to the proposition "there is x such that x is a c and x is j". Note that a logical formulation is a semantic formulation of a proposition, which is that proposition in the case of existential propositions. I agree with Terry that existential propositions are distinguished. Note that the existential proposition uses three of the four Frege-Russell distinctions between different meanings of 'is': existence, generic implication (inclusion), and identity. The fourth meaning of 'is' is predication. Note that if the above state of affairs obtains, the state of affairs is an actuality, i.e. the actual world includes the state of affairs. If the state of affairs does not obtain, some other possible world includes it. The existential proposition is true just in each possible world that includes the corresponding state of affairs. Stan On Sep 18, 2011, at 4:54 PM, Terry Halpin wrote: Hi Ron <> I distinguish purely existential propositions (e.g. there is a country code 'US') from propositions that predicate over individuals (e.g. Ron was born in the USA). To me, a proposition is what it is that is asserted by a declarative sentence (or the declarative component of a sentence), and is always true or false (but not both). The SBVR "meaning that is true or false" seems to correspond pretty close to this. I don't consider a variable to be a meaning. Cheers Terry On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross wrote: At 05:00 PM 9/18/2011, Terry Halpin wrote: <> I disagree with Stan here. I use "role" in two ways. A "fact role" is a part played in a fact, and always involves a logical predicate (e.g. The fact type "Person drives Car" has two roles, one being the role of driving which is played by a person, and the other being the role of being driven which is played by a car). A "role type" is a non-rigid type (e.g. Student), unlike rigid types such as Person or Thing. The elementary fact "John Hall is alive" involves an individual and a unary predicate "is alive", which may be treated as the role of being alive. The existential fact "John Hall exists" involves an individual and an existential quantifier. The usual view in logic, with which I concur, is that existence is not a predicate (and certainly not a first-order predicate). "Exists" is not the same kind of construct as "is alive", and it does not involve a role. If we use the individual constant "j" for John Hall, then "John Hall is alive" may be formalized as "j isAlive" (in postfix notation), but "John Hall exists" is formalized as "Exists x x = j". Terry, Would it be fair to say then that there are two categories of proposition (relevant to SBVR) -- (1) existential propositions, and (2) propositions involving some role? The current definition of "proposition:" in SBVR seems inadequate ... "meaning that is true or false " ... and this is one source of much consternation. For example, that definition would seem to admit boolean variables. I think we really need something better. Thanks, Ron Cheers Terry On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx wrote: Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. >>What about existential propositions? e.g.., John Hall exists. These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), pp. 25-50 Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. If a fact model meets the truth conditions, the proposition is true for that fact model, for the possible world modeled by the fact model. What it means for a fact model to meet the conditions of a proposition is spelled out in the model-theoretic semantics of SBVR, which, as Ed pointed out, is not well documented. This is an area for future work on SBVR. The model theory of a language is a significant part of it's formal grounding, important to interpreting the language and getting translations correct. Not all 'artificial' languages have a model theory, e.g. UML does not. Stan On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" wrote: >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info From: "Barkmeyer, Edward J" To: Terry Halpin , "Ronald G. Ross" CC: Stan Hendryx , "sbvr-rtf@omg.org" , "date-time@omg.org" Date: Mon, 19 Sep 2011 17:15:38 -0400 Subject: RE: meaning of 'proposition' Thread-Topic: meaning of 'proposition' Thread-Index: Acx2XjsaS4qN7fy/RAC/QME20tyO9wAscxgi Accept-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US X-MIME-Autoconverted: from quoted-printable to 8bit by amethyst.omg.org id p8JLGA5C007276 Terry wrote: > To me, a proposition is what it is that is asserted by a declarative sentence (or the declarative component of a sentence), > and is always true or false (but not both). The SBVR "meaning that is true or false" seems to correspond pretty close to this. Well, the 'but not both' is precisely the point of the issue. In 'John is in London at least once a month', the SBVR approach to modeling this statement makes 'John is in London' a declarative component of a sentence, but possibly not what Terry means by 'the declarative component of a sentence'. If the same fact model includes 'John is not in London on Fridays', it is clear that the component 'John is in London' cannot be either true or false, unless it is both. In either case, it fails Terry's proposed definition of 'proposition' and therefore is not a proposition. It follows that it does not correspond to any 'state of affairs'. This is the dilemma that motivated the issue. The particular problem here is that Terry's definition is intended to relate the concept 'proposition' only to sentences in their entirety, i.e., John is in London at least once a month is a proposition, but no subset of it is, or at least, no subset is being asked whether it is. That resolves the issue, but with the same consequence: 'John is in London' is not a declarative sentence in the fact model, and therefore does not represent a proposition, and thus cannot correspond to a state of affairs. So, it doesn't support the desired SBVR semantics either. -Ed -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 ________________________________________ From: Terry Halpin [terry.halpin@logicblox.com] Sent: Sunday, September 18, 2011 7:54 PM To: Ronald G. Ross Cc: Stan Hendryx; Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org Subject: Re: meaning of 'proposition' Hi Ron <> I distinguish purely existential propositions (e.g. there is a country code 'US') from propositions that predicate over individuals (e.g. Ron was born in the USA). To me, a proposition is what it is that is asserted by a declarative sentence (or the declarative component of a sentence), and is always true or false (but not both). The SBVR "meaning that is true or false" seems to correspond pretty close to this. I don't consider a variable to be a meaning. Cheers Terry On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross > wrote: At 05:00 PM 9/18/2011, Terry Halpin wrote: <> I disagree with Stan here. I use "role" in two ways. A "fact role" is a part played in a fact, and always involves a logical predicate (e.g. The fact type "Person drives Car" has two roles, one being the role of driving which is played by a person, and the other being the role of being driven which is played by a car). A "role type" is a non-rigid type (e.g. Student), unlike rigid types such as Person or Thing. The elementary fact "John Hall is alive" involves an individual and a unary predicate "is alive", which may be treated as the role of being alive. The existential fact "John Hall exists" involves an individual and an existential quantifier. The usual view in logic, with which I concur, is that existence is not a predicate (and certainly not a first-order predicate). "Exists" is not the same kind of construct as "is alive", and it does not involve a role. If we use the individual constant "j" for John Hall, then "John Hall is alive" may be formalized as "j isAlive" (in postfix notation), but "John Hall exists" is formalized as "Exists x x = j". Terry, Would it be fair to say then that there are two categories of proposition (relevant to SBVR) -- (1) existential propositions, and (2) propositions involving some role? The current definition of "proposition:" in SBVR seems inadequate ... "meaning that is true or false " ... and this is one source of much consternation. For example, that definition would seem to admit boolean variables. I think we really need something better. Thanks, Ron Cheers Terry On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx > wrote: Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. >>What about existential propositions? e.g.., John Hall exists. These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), pp. 25-50 Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. If a fact model meets the truth conditions, the proposition is true for that fact model, for the possible world modeled by the fact model. What it means for a fact model to meet the conditions of a proposition is spelled out in the model-theoretic semantics of SBVR, which, as Ed pointed out, is not well documented. This is an area for future work on SBVR. The model theory of a language is a significant part of it's formal grounding, important to interpreting the language and getting translations correct. Not all 'artificial' languages have a model theory, e.g. UML does not. Stan On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" > wrote: >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. >From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" < mailto:rross@brsolutions.com>> wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < mailto:keri_ah@mac.com> < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info Blog || http://www.RonRoss.info/blog/ LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 Twitter || https://twitter.com/Ronald_G_Ross Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 305022.8568.bm@omp1013.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316477874; bh=mX/cR6b6OCb7ANeo9MqDkaT/x/kv3Vx9lWSntKh+CSs=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:From:To:Cc:References:Subject:Date:Organization:Message-ID:MIME-Version:Content-Type:Content-Transfer-Encoding:X-Mailer:Thread-Index:X-MimeOLE:In-Reply-To; b=wKCdvGNzctfUw/BGBoLn4U//+EnrBTjQiQKGtO4yApSJ2fSxJZVirMsKFU3SpquTd4lIASqa2mSbCDFUsAW5JzpMQGsva4bhskP+DbLY5P6ud5OEb0adNzxe+iibuzakyVJ8mr0VF1Kpl6B6R0SAzHBv7ZF4UHGZP2BCLO+7YRo= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: HvOLlLMVM1n8ZJWcqzmloU3zftAdvJxOuzVO4vBDUXZkIVy x7fhVHADZ7odgpSJVeEOqOq07Bg1Xo_CjCM6.fCDFFQesp_ZnjLwNreyL9xz lJXJaT8MGOLvQdN3J.ogPpGPh.FaudXiEiu5nKoOpYsW_i.ykhqrioyo7r2k KTBfRPgtDBx0jQQD.UeP8EN_a0CL0FN2qs0m1bClGt2RfJ0WxmVvXDDOhJUR uNXWCnjNsCFVZHdtXWpjwGvz4GtjTgU1ACLCp3oPhEYC7XudDBom5TtxenUc L7_4tzCfV5U7MALjugi_6dslSSgWlNBQpg4xbLIaP0zprAHiVbCzbiayj.g8 sOcz16.cXlpVo4FfLHA_KHVzmYcTv6HrBc9KRoFf6L6gTmKKTqETfWNAXmTK krzQ_jXA1iC927A5tcNni8Rsl7gy7dbgO9faX3abcpLXH X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI From: "Stan Hendryx" To: "'Barkmeyer, Edward J'" , "'Terry Halpin'" , "'Ronald G. Ross'" Cc: , Subject: RE: meaning of 'proposition' Date: Mon, 19 Sep 2011 17:17:51 -0700 Organization: Hendryx & Associates X-Mailer: Microsoft Office Outlook 11 Thread-Index: Acx2XjsaS4qN7fy/RAC/QME20tyO9wAscxgiAATWTLA= The matter of "always true or false (but not both)" is explained by considering that states of affairs denoted by sentences like 'John is in London at least once a month' include the state of affairs denoted by 'John is in London', where the latter is interpreted as being the primitive notion of John being in London, regardless of time. In this context, 'is' does not mean John is in London now, just what it means for John to be in London at all. It is this primitive state of affairs that is said to occur at least once a month, but saying it recurs is just a figure of speech. Formally, such a state of affairs does not literally itself recur; no state of affairs happens more than once. Rather, the primitive state of affairs is included in each state of affairs that is a monthly occurrence. The primitive is what occurs in an occurrence. In Plantinga's theory of actualism, the primitive states of affairs are called "haecceities", which is the essential property of a thing being just what it is. 'Includes' is transitive: if A includes B and B includes C, then A includes C. Thus, the graph of the 'includes' relation in a possible world is a lattice, not a tree, because the primitives can be part of many particular occurrences. The lattice is defined by the set of states of affairs the possible world includes (the set is called the domain of the possible world) and the 'includes' relation. The possible world is the top of the lattice, so it is a sort of maximal state of affairs, being the greatest lower bound of the set under 'includes'. A possible world includes a particular state of affairs at most once, as a member of its domain. The primitive proposition 'John is in London' is true in a possible world just in case the possible world includes at least one state of affairs that includes the primitive state of affairs that John is in London. The primitive is a genuine state of affairs; it occurs for a time interval that is the convex hull of each state of affairs that includes it. Stan > -----Original Message----- > From: Barkmeyer, Edward J [mailto:edward.barkmeyer@nist.gov] > Sent: Monday, September 19, 2011 2:16 PM > To: Terry Halpin; Ronald G. Ross > Cc: Stan Hendryx; sbvr-rtf@omg.org; date-time@omg.org > Subject: RE: meaning of 'proposition' > > Terry wrote: > > > To me, a proposition is what it is that is asserted by a declarative > sentence (or the declarative component of a sentence), > > and is always true or false (but not both). The SBVR "meaning that is > true or false" seems to correspond pretty close to this. > > Well, the 'but not both' is precisely the point of the issue. In 'John is > in London at least once a month', the SBVR approach to modeling this > statement makes 'John is in London' a declarative component of a sentence, > but possibly not what Terry means by 'the declarative component of a > sentence'. If the same fact model includes 'John is not in London on > Fridays', it is clear that the component 'John is in London' cannot be > either true or false, unless it is both. In either case, it fails Terry's > proposed definition of 'proposition' and therefore is not a proposition. > It follows that it does not correspond to any 'state of affairs'. This is > the dilemma that motivated the issue. > > The particular problem here is that Terry's definition is intended to > relate the concept 'proposition' only to sentences in their entirety, i.e., > John is in London at least once a month is a proposition, but no subset of > it is, or at least, no subset is being asked whether it is. That resolves > the issue, but with the same consequence: 'John is in London' is not a > declarative sentence in the fact model, and therefore does not represent a > proposition, and thus cannot correspond to a state of affairs. So, it > doesn't support the desired SBVR semantics either. > > -Ed > > -- > Edward J. Barkmeyer Email: edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 > Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 > ________________________________________ > From: Terry Halpin [terry.halpin@logicblox.com] > Sent: Sunday, September 18, 2011 7:54 PM > To: Ronald G. Ross > Cc: Stan Hendryx; Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org > Subject: Re: meaning of 'proposition' > > Hi Ron > > < categories of proposition (relevant to SBVR) -- (1) existential > propositions, and (2) propositions involving some role? > > The current definition of "proposition:" in SBVR seems inadequate ... > "meaning thatis true or false " ... and this is one source of much > consternation. For example, that definition would seem to admit boolean > variables. I think we really need something better.>> > > I distinguish purely existential propositions (e.g. there is a country > code 'US') from propositions that predicate over individuals (e.g. Ron was > born in the USA). To me, a proposition is what it is that is asserted by a > declarative sentence (or the declarative component of a sentence), and is > always true or false (but not both). The SBVR "meaning that is true or > false" seems to correspond pretty close to this. I don't consider a > variable to be a meaning. > > Cheers > Terry > > > > On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross > > wrote: > At 05:00 PM 9/18/2011, Terry Halpin wrote: > < > Stan wrote: These, too, involve roles. "John Hall exists" is a statement > that represents an instance of the unary fact type 'thing exists'. The > role is 'thing'. The bindable target of the role in the example statement > is "John Hall". The actuality denoted by the statement is John Hall. Note > that SBVR does not include the fact type 'statement denotes state of > affairs', so, using SBVR terms, we would say "the actuality that > corresponds to the proposition expressed by the statement is John Hall." > >> > > I disagree with Stan here. I use "role" in two ways. A "fact role" is a > part played in a fact, and always involves a logical predicate (e.g. The > fact type "Person drives Car" has two roles, one being the role of driving > which is played by a person, and the other being the role of being driven > which is played by a car). A "role type" is a non-rigid type (e.g. > Student), unlike rigid types such as Person or Thing. > > The elementary fact "John Hall is alive" involves an individual and a > unary predicate "is alive", which may be treated as the role of being > alive. The existential fact "John Hall exists" involves an individual and > an existential quantifier. The usual view in logic, with which I concur, > is that existence is not a predicate (and certainly not a first-order > predicate). "Exists" is not the same kind of construct as "is alive", and > it does not involve a role. If we use the individual constant "j" for John > Hall, then "John Hall is alive" may be formalized as "j isAlive" (in > postfix notation), but "John Hall exists" is formalized as "Exists x x = > j". > > Terry, > > Would it be fair to say then that there are two categories of > proposition (relevant to SBVR) -- (1) existential propositions, and (2) > propositions involving some role? > > The current definition of "proposition:" in SBVR seems > inadequate ... "meaning that is true or false " ... and this is one source > of much consternation. For example, that definition would seem to admit > boolean variables. I think we really need something better. > > Thanks, > Ron > > > Cheers > Terry > > On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx > > wrote: > > Ron wrote: > Why would we want to get into 'artificial' languages ... *especially* ones > that are already formally grounded?? > > Machines use different formal languages, so there is a need to translate > between these languages for machines to communicate just as there is a > need to translate between natural languages business people use. > > >>What about existential propositions? e.g.., John Hall exists. > > These, too, involve roles. "John Hall exists" is a statement that > represents an instance of the unary fact type 'thing exists'. The role is > 'thing'. The bindable target of the role in the example statement is "John > Hall". The actuality denoted by the statement is John Hall. Note that SBVR > does not include the fact type 'statement denotes state of affairs', so, > using SBVR terms, we would say "the actuality that corresponds to the > proposition expressed by the statement is John Hall." > > In his landmark paper of 1892, Frege explained that a sign has both a > reference and a sense. In SBVR, I take a statement to be a sign, a state > of affairs to be the reference (denotation) of the sign, and a proposition > to be the sense (meaning) of the sign. I thought it might interest you to > see some of Frege's own words on this (from Max Black's 1960 translation > of the original German): > > "It is natural, now, to think of there being connected with a sign (name, > combination of words, letter), besides that to which the sign refers, > which may be called the reference of the sign, also what I should like to > call the sense of the sign, wherein the mode of presentation is contained. > In our example,...the reference of 'evening star' would be the same as > that of 'morning star', but not the sense. > > "The regular connection between a sign, its sense, and its reference is of > such a kind that to the sign there corresponds a definite sense and to > that in turn a definite reference, while to a given reference (an object) > there does not belong only a single sign. The same sense has different > expressions in different languages or even in the same language...It may > be granted that every grammatically well-formed expression representing a > proper name always has a sense. But this is not to say that to the sense > there also corresponds a reference. The words 'the celestial body most > distant from the Earth' have a sense, but it is very doubtful if they also > have a reference...In grasping a sense, one is not certainly assured of a > reference. > [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) > and says that every grammatically well-formed expression that is not > paradoxical and that does not deny a necessity is certainly assured of a > reference (a state of affairs) in some possible world. If the reference is > included in, is part of, the actual world, the state of affairs is an > actuality. This is not exactly the SBVR notion of an actuality.] > > "If words are used in the ordinary way, what one intends to speak of is > their reference. It can also happen, however, that one wishes to talk > about the words themselves or their sense. This happens, for instance, > when the words of another are quoted..." > --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in > Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), > pp. 25-50 > > Thanks, Ron, for the clarification about the context you mentioned. I > understand from your comment that the idea of a proposition giving truth > conditions is new in the SBVR discussions. If a fact model meets the truth > conditions, the proposition is true for that fact model, for the possible > world modeled by the fact model. What it means for a fact model to meet > the conditions of a proposition is spelled out in the model-theoretic > semantics of SBVR, which, as Ed pointed out, is not well documented. This > is an area for future work on SBVR. The model theory of a language is a > significant part of it's formal grounding, important to interpreting the > language and getting translations correct. Not all 'artificial' languages > have a model theory, e.g. UML does not. > > Stan > > On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" > > wrote: > > >>Good discussion. Some notes inserted like this below. > > Ron > > At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: > Stan Hendryx wrote: > Ed wrote: > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, > > Three SBVR fact types are particularly relevant here: > statement expresses proposition > closed logical formulation formalizes statement > closed logical formulation means proposition > > Only in passing. The proposition (meaning) is the same, whether it is > expressed in English or Structured English or Chinese, or SBVR clause 9 > XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR > is that it doesn't seem to have considered that all of these are > Statements -- expressions representing propositions -- and several of them > are at least as formally grounded as SBVR aspires to be. > > These are related by this rule, which is a composite of several SBVR > necessities: > Each statement expresses exactly one proposition and the statement is > formalized by a closed logical formulation that means just that > proposition. > > A statement in CLIF doesn't need to be formalized by an SBVR closed > logical formulation. It is formalized by the language. Statement means > 'representation of a proposition' -- any representation. > > This should be qualified by saying "each statement that is not paradoxical > expresses...", since paradoxes are nonsensical, not having a meaning, no > logical formulation, e.g. "this sentence is false." Such sentences do not > express propositions, either. > > A closed logical formulation is a representation, a translation, of the > statement it formalizes into the language of predicate logic. > > Well, no. It formalizes it into a language defined in clause 9, 13 and 15 > whose semantics is said to be a dialect of predicate logic. There is no > 'the language of predicate logic'; there are many. And whether the logic > language that Clause 9 introduces has a well-defined formal semantics is > still debatable. None has ever been formally written down. > > A CLIF representation of a proposition is a representation in a standard > language for predicate logic. And you yourself used Quine's mathematical > notation to state propositions. Surely that didn't need to be formalized > in SBVR LRMV to be formalized in the language of predicate logic. > > This is all about thinking that Statement means 'natural language > statement' or perhaps 'Structured English statement', and that idea > pervades SBVR. In Date/Time, we have expressed most of the formal > propositions in SBVR SE, OCL and CLIF, of which the last two are formal > and have formal semantics. In point of fact, we have no tool that > generates the SBVR LRMV closed logical formulations. But hope springs > eternal, at least outside of NIST. > > Ron wrote: > "Truth conditions" is new to me in this context. > > This idea came from G. Frege ca. 1892. His work, which includes the > invention of predicate calculus, is the foundation most subsequent work in > logic and formal semantics. > > >>Next time when I say "this context" I will take the time to be more > clear for you (Stan). "This context" means the SBVR-related e-mails I have > been recently reading regarding the definition of "proposition" in SBVR. > There, better? > > > > Certainly logical formulations should be considered out of bounds for > defining what a proposition is. > > On the contrary, a closed logical formulation that means a proposition > definitively says what the proposition is. SBVR gives a definition of the > concept 'proposition': meaning that is true or false. > > I agree with Ron. The point of all this is that the issue of what a > proposition is does not depend on how it is represented. > > >>Thank you, Ed. That's so obvious it should speak for itself (but of > course it can't). > > > And it certainly doesn't depend on its being represented in an XML form > which no tools generate. 'John is in London' definitively says what the > proposition is, at least to speakers of English. You cannot express it > any more clearly in predicate logic, because that would also require that > you explain that the symbol 'John' is an individual concept (logical > constant) that refers to a specific person, and similarly 'London', and > that the predicate symbol 'lives in' refers to the verb concept that is > designated 'lives in' in English. The formal language is just a grammar > that eliminates certain ambiguities in the interpretation, assuming that > you know what the terminal symbols mean. > > To understand the relationship between fact types and their roles, and > propositions, consider that an expression of an instance of a fact type is > a sentence, an atomic sentence. For example, the sentence "Stan was born > in the US," expresses an instance of the binary fact type 'person was born > in country'. Such a sentence expresses a proposition. The sentence is > formalized by an atomic formulation with role bindings, i.e. it is a > closed logical formulation; I call it a closed atomic formulation (my term, > not SBVR). Each role binding is of a role of the fact type. Each role > binding binds to a bindable target, which is either a bound variable, an > individual (Stan and the US in the example) or an identifying expression. > The closed atomic formulation means the proposition that is expressed by > the sentence. The proposition corresponds to a state of affairs. The state > of affairs is the instance of the fact type. If the state of affairs is an > actuality, the proposition is true. To sum up, an instance of a fact type > is a state of affairs that corresponds to the proposition that is > expressed by the statement of the instance of the fact type. > > I thought SBVR was clearer on this. An instance of a fact type is an > actuality that is understood as involving some things in the roles > specified in the fact type form(s). > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? > > If the statement of the proposition is formal in the SBVR sense (colored), > then the proposition will always involve some role, > > I agree. "Murphy's Law" is an expression that denotes a proposition, and > that exrpression does not involve any thing in any role. But the meaning > that is "Murphy's Law" depends on an intransitive verb concept and > involves the general pronoun 'anything' in the sole role in the unary fact > type 'thing goes wrong'. That is one more reason to stop worrying about > the nature of the expression. > The nature of the proposition as a meaning is to conceptualize a situation > as involving things in roles. I know of no counterexamples. > >>What about existential propositions? e.g.., John Hall exists. ... Some > person exists. > > From a business (not logician's) point of view, these to me to involve > roles too. So I agree ... I can think of no counterexamples. Can anyone > else? > > If not, I believe this is a fundamental characteristic in the meaning of > 'proposition'. It seems right at the heart of what SBVR is about. > > > -Ed > > since every closed logical formulation involves at least one fact type > (verb concept) and it's atomic formulation. We know this because every > statement has a verb and by hypothesis, the verb is formalized. In SBVR, > verbs are formalized as fact type designations, and each fact type has at > least one role. > Stan > > On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > < > mailto:rross@brsolutions.com>> wrote: > > At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: > > > Stan Hendryx wrote: > Keri wrote: > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > Yes. The meaning of a proposition is its truth conditions, what will make > it true or false. These conditions are stated precisely by the closed > logical formulation that formalizes a statement that expresses the > proposition. Truth conditions are independent of any facts. > > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, Stan's position is the basis > for the issue. The nature of the proposition is not that it has a truth > value but rather that it is a set of conditions that determines a truth > value in a given world. The issue as written described that set of > conditions as the 'conceptualization of a situation', but perhaps Stan's > characterization is a bit clearer. Those conditions involve things > playing roles (which, according to SBVR is the nature of a state of > affairs). > The truth value of a proposition, it being true or false, depends on the > facts in each fact model in which it is evaluated. It may be true in some > fact models and false in others. It may be unknown. We know what a > proposition means even if it is false or even if its truth value is > unknown. > I hope this helps clarify what Ron meant. > > Stan is probably right that I misunderstood Ron's point. Ron apparently > meant that the meaning that is a proposition is independent of its truth > value. > > Yes, I meant: By definition, a meaning that is a proposition is > independent of its truth value. > > We fully agree. Does Ron agree with Stan that the nature of the meaning > is the 'truth conditions'? And if we agree, can we write a resolution to > 16526 to say that? > > "Truth conditions" is new to me in this context. Certainly logical > formulations should be considered out of bounds for defining what a > proposition is. Ed, you raised some interesting points in your earlier e- > mail response about roles and fact types with respect to defining > propositions I want to hear discussion about. > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? I say > cautiously because (a) I don't know where that takes things, and (b) I > haven't checked to see if the definitions would become circular. > > Ron > > > -Ed > > > Stan > > > > On Sep 16, 2011, at 10:59 AM, keri > < mailto:keri_ah@mac.com> < < > mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: > > Mark, > > Here are some of the points I captured in my notes. (I don't show that > the bits about "independent of any Universe of Discourse" was part of what > Ron said, but perhaps so and I missed that. From the discussion around > the example proposition statements it did appear that they pertained to > some world/UOD.) > > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > I also captured points such as: > * Being true or false does not depend on (a) if it's known that > it's true or false or (b) that it's interesting that it's true > or false. > * Being true or false is different from ambiguity. > > In regards to the example "John is in London." I have to conclude that the > reason it is a proposition (without knowing whether or not John is > actually in London) is because this statement (of proposition) is /based > on/ the vocabulary's fact type "person is in place" and the further aspect > that a person must be in exactly one place. If this is the meaning that > is understood then it can be seen that any statement about some person > being in some place must be either true or false, without knowing which it > is. (I think that's how it works.) When we exchange examples > propositions we have to have the meaning come from the vocabulary concepts > that the proposition is /based on/. Otherwise, there is no meaning. > > - Keri > > > On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: > > During today's RTF discussion, we discussed this example from Ed's note of > September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My > purpose in writing this email is to summarize what I think I heard. > Please comment if you think I got it wrong. > > > Given a fact model comprising: John is in London during 5-8 September > > 2011. John is in St. Imier during 9-10 September 2011. The time of > > interest in the Universe of Discourse is Sept 5-10, 2011. The > > proposition expressed by 'John is in London' is used in representing > > this fact model, both in SBVR examples, and formally in Date/Time. Is > > it true or false? I believe that in that actual world, the proposition > > is neither true nor false. Therefore, it is a counterexample to what > > Don says. > > 1. SBVR's definition of "proposition" is "meaning that is true or false". > Despite the use of the word "is" in the definition, this definition does > not mean that you have to interpret an expression against a Universe of > Discourse to determine whether the expression expresses a proposition. > Ron Ross said something to the effect that it is very important that a > proposition is one independent of any Universe of Discourse. I think he's > right -- but the current definition is not clear about that. I understand > that the RTF will propose a change to clarify this point. > > 2. Regarding the example given above, what I heard is that whether "John > is in London" may not be known in the model, and is not inferable solely > from the two facts given. So the expression may not be decidable, but it > is still expresses a proposition. > > > Is this a correct summary? > -------------------------------- > Mark H. Linehan > STSM, Model Driven Business Transformation > IBM Research > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov < mailto:edbark@nist.gov> > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > [Ronald G. Ross Contact Information] > http://www.brsolutions.com/email/wordpress-2.png *Blog* || > < http://www.ronross.info/blog/ > > http://www.RonRoss.info/blog/ > http://www.brsolutions.com/email/linkedin.png *LinkedIn* || > < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.brsolutions.com/email/twitter.png < > https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> > *Twitter* || > < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross > http://www.brsolutions.com/email/button-green.png *Homepage* > || < http://www.ronross.info/ > > http://www.RonRoss.info > > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 > Twitter || https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > Twitter || > https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > From: "Barkmeyer, Edward J" To: Stan Hendryx CC: "sbvr-rtf@omg.org" , "date-time@omg.org" Date: Tue, 20 Sep 2011 11:35:27 -0400 Subject: RE: meaning of 'proposition' Thread-Topic: meaning of 'proposition' Thread-Index: Acx11ua+mEC2cJR2SOWNwkMfQq4hRgB0R7M3 Accept-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US X-MIME-Autoconverted: from quoted-printable to 8bit by amethyst.omg.org id p8KFZsqM003938 Stan wrote: [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] It is a lot more that 'not exactly'. This model is importantly different from the SBVR idea that there exists a referent of the grammatically well-formed expression -- a corresponding state of affairs -- in every possible world, and the expression may even deny a necessity. We can talk about impossible situations; therefore they exist. But Terry claims to be consistent with Kripke semantics; so some part of this reading is inaccurate. This is why we the amateur logicians need to stop debating this stuff, find an expert or two, and lay our problem -- carefully phrased -- before them. -Ed -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 ________________________________________ From: Stan Hendryx [stan@hendryxassoc.com] Sent: Sunday, September 18, 2011 3:49 AM To: Ronald G. Ross Cc: Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org Subject: Re: meaning of 'proposition' Ron wrote: Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. >>What about existential propositions? e.g.., John Hall exists. These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), pp. 25-50 Thanks, Ron, for the clarification about the context you mentioned. I understand from your comment that the idea of a proposition giving truth conditions is new in the SBVR discussions. If a fact model meets the truth conditions, the proposition is true for that fact model, for the possible world modeled by the fact model. What it means for a fact model to meet the conditions of a proposition is spelled out in the model-theoretic semantics of SBVR, which, as Ed pointed out, is not well documented. This is an area for future work on SBVR. The model theory of a language is a significant part of it's formal grounding, important to interpreting the language and getting translations correct. Not all 'artificial' languages have a model theory, e.g. UML does not. Stan On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" > wrote: >>Good discussion. Some notes inserted like this below. Ron At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: Stan Hendryx wrote: Ed wrote: Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Three SBVR fact types are particularly relevant here: statement expresses proposition closed logical formulation formalizes statement closed logical formulation means proposition Only in passing. The proposition (meaning) is the same, whether it is expressed in English or Structured English or Chinese, or SBVR clause 9 XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR is that it doesn't seem to have considered that all of these are Statements -- expressions representing propositions -- and several of them are at least as formally grounded as SBVR aspires to be. These are related by this rule, which is a composite of several SBVR necessities: Each statement expresses exactly one proposition and the statement is formalized by a closed logical formulation that means just that proposition. A statement in CLIF doesn't need to be formalized by an SBVR closed logical formulation. It is formalized by the language. Statement means 'representation of a proposition' -- any representation. This should be qualified by saying "each statement that is not paradoxical expresses...", since paradoxes are nonsensical, not having a meaning, no logical formulation, e.g. "this sentence is false." Such sentences do not express propositions, either. A closed logical formulation is a representation, a translation, of the statement it formalizes into the language of predicate logic. Well, no. It formalizes it into a language defined in clause 9, 13 and 15 whose semantics is said to be a dialect of predicate logic. There is no 'the language of predicate logic'; there are many. And whether the logic language that Clause 9 introduces has a well-defined formal semantics is still debatable. None has ever been formally written down. A CLIF representation of a proposition is a representation in a standard language for predicate logic. And you yourself used Quine's mathematical notation to state propositions. Surely that didn't need to be formalized in SBVR LRMV to be formalized in the language of predicate logic. This is all about thinking that Statement means 'natural language statement' or perhaps 'Structured English statement', and that idea pervades SBVR. In Date/Time, we have expressed most of the formal propositions in SBVR SE, OCL and CLIF, of which the last two are formal and have formal semantics. In point of fact, we have no tool that generates the SBVR LRMV closed logical formulations. But hope springs eternal, at least outside of NIST. Ron wrote: "Truth conditions" is new to me in this context. This idea came from G. Frege ca. 1892. His work, which includes the invention of predicate calculus, is the foundation most subsequent work in logic and formal semantics. >>Next time when I say "this context" I will take the time to be more clear for you (Stan). "This context" means the SBVR-related e-mails I have been recently reading regarding the definition of "proposition" in SBVR. There, better? Certainly logical formulations should be considered out of bounds for defining what a proposition is. On the contrary, a closed logical formulation that means a proposition definitively says what the proposition is. SBVR gives a definition of the concept 'proposition': meaning that is true or false. I agree with Ron. The point of all this is that the issue of what a proposition is does not depend on how it is represented. >>Thank you, Ed. That's so obvious it should speak for itself (but of course it can't). And it certainly doesn't depend on its being represented in an XML form which no tools generate. 'John is in London' definitively says what the proposition is, at least to speakers of English. You cannot express it any more clearly in predicate logic, because that would also require that you explain that the symbol 'John' is an individual concept (logical constant) that refers to a specific person, and similarly 'London', and that the predicate symbol 'lives in' refers to the verb concept that is designated 'lives in' in English. The formal language is just a grammar that eliminates certain ambiguities in the interpretation, assuming that you know what the terminal symbols mean. To understand the relationship between fact types and their roles, and propositions, consider that an expression of an instance of a fact type is a sentence, an atomic sentence. For example, the sentence "Stan was born in the US," expresses an instance of the binary fact type 'person was born in country'. Such a sentence expresses a proposition. The sentence is formalized by an atomic formulation with role bindings, i.e. it is a closed logical formulation; I call it a closed atomic formulation (my term, not SBVR). Each role binding is of a role of the fact type. Each role binding binds to a bindable target, which is either a bound variable, an individual (Stan and the US in the example) or an identifying expression. The closed atomic formulation means the proposition that is expressed by the sentence. The proposition corresponds to a state of affairs. The state of affairs is the instance of the fact type. If the state of affairs is an actuality, the proposition is true. To sum up, an instance of a fact type is a state of affairs that corresponds to the proposition that is expressed by the statement of the instance of the fact type. I thought SBVR was clearer on this. An instance of a fact type is an actuality that is understood as involving some things in the roles specified in the fact type form(s). I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? If the statement of the proposition is formal in the SBVR sense (colored), then the proposition will always involve some role, I agree. "Murphy's Law" is an expression that denotes a proposition, and that exrpression does not involve any thing in any role. But the meaning that is "Murphy's Law" depends on an intransitive verb concept and involves the general pronoun 'anything' in the sole role in the unary fact type 'thing goes wrong'. That is one more reason to stop worrying about the nature of the expression. The nature of the proposition as a meaning is to conceptualize a situation as involving things in roles. I know of no counterexamples. >>What about existential propositions? e.g.., John Hall exists. ... Some person exists. >From a business (not logician's) point of view, these to me to involve roles too. So I agree ... I can think of no counterexamples. Can anyone else? If not, I believe this is a fundamental characteristic in the meaning of 'proposition'. It seems right at the heart of what SBVR is about. -Ed since every closed logical formulation involves at least one fact type (verb concept) and it's atomic formulation. We know this because every statement has a verb and by hypothesis, the verb is formalized. In SBVR, verbs are formalized as fact type designations, and each fact type has at least one role. Stan On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" < mailto:rross@brsolutions.com>> wrote: At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: Stan Hendryx wrote: Keri wrote: What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. Yes. The meaning of a proposition is its truth conditions, what will make it true or false. These conditions are stated precisely by the closed logical formulation that formalizes a statement that expresses the proposition. Truth conditions are independent of any facts. Ignoring the idea that a proposition has to be stated as an SBVR closed logical formulation in order to be expressed, Stan's position is the basis for the issue. The nature of the proposition is not that it has a truth value but rather that it is a set of conditions that determines a truth value in a given world. The issue as written described that set of conditions as the 'conceptualization of a situation', but perhaps Stan's characterization is a bit clearer. Those conditions involve things playing roles (which, according to SBVR is the nature of a state of affairs). The truth value of a proposition, it being true or false, depends on the facts in each fact model in which it is evaluated. It may be true in some fact models and false in others. It may be unknown. We know what a proposition means even if it is false or even if its truth value is unknown. I hope this helps clarify what Ron meant. Stan is probably right that I misunderstood Ron's point. Ron apparently meant that the meaning that is a proposition is independent of its truth value. Yes, I meant: By definition, a meaning that is a proposition is independent of its truth value. We fully agree. Does Ron agree with Stan that the nature of the meaning is the 'truth conditions'? And if we agree, can we write a resolution to 16526 to say that? "Truth conditions" is new to me in this context. Certainly logical formulations should be considered out of bounds for defining what a proposition is. Ed, you raised some interesting points in your earlier e-mail response about roles and fact types with respect to defining propositions I want to hear discussion about. I ask this very. very cautiously: Can anybody think of an example proposition in the SBVR sense that wouldn't involve some role? I say cautiously because (a) I don't know where that takes things, and (b) I haven't checked to see if the definitions would become circular. Ron -Ed Stan On Sep 16, 2011, at 10:59 AM, keri < mailto:keri_ah@mac.com> < < mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: Mark, Here are some of the points I captured in my notes. (I don't show that the bits about "independent of any Universe of Discourse" was part of what Ron said, but perhaps so and I missed that. From the discussion around the example proposition statements it did appear that they pertained to some world/UOD.) What I got for Ron's statement was: It's an important feature of SBVR that a proposition is what it is, *independent of knowing any facts*. I also captured points such as: * Being true or false does not depend on (a) if it's known that it's true or false or (b) that it's interesting that it's true or false. * Being true or false is different from ambiguity. In regards to the example "John is in London." I have to conclude that the reason it is a proposition (without knowing whether or not John is actually in London) is because this statement (of proposition) is /based on/ the vocabulary's fact type "person is in place" and the further aspect that a person must be in exactly one place. If this is the meaning that is understood then it can be seen that any statement about some person being in some place must be either true or false, without knowing which it is. (I think that's how it works.) When we exchange examples propositions we have to have the meaning come from the vocabulary concepts that the proposition is /based on/. Otherwise, there is no meaning. - Keri On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: During today's RTF discussion, we discussed this example from Ed's note of September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My purpose in writing this email is to summarize what I think I heard. Please comment if you think I got it wrong. > Given a fact model comprising: John is in London during 5-8 September > 2011. John is in St. Imier during 9-10 September 2011. The time of > interest in the Universe of Discourse is Sept 5-10, 2011. The > proposition expressed by 'John is in London' is used in representing > this fact model, both in SBVR examples, and formally in Date/Time. Is > it true or false? I believe that in that actual world, the proposition > is neither true nor false. Therefore, it is a counterexample to what > Don says. 1. SBVR's definition of "proposition" is "meaning that is true or false". Despite the use of the word "is" in the definition, this definition does not mean that you have to interpret an expression against a Universe of Discourse to determine whether the expression expresses a proposition. Ron Ross said something to the effect that it is very important that a proposition is one independent of any Universe of Discourse. I think he's right -- but the current definition is not clear about that. I understand that the RTF will propose a change to clarify this point. 2. Regarding the example given above, what I heard is that whether "John is in London" may not be known in the model, and is not inferable solely from the two facts given. So the expression may not be decidable, but it is still expresses a proposition. Is this a correct summary? -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research -- Edward J. Barkmeyer Email: edbark@nist.gov < mailto:edbark@nist.gov> National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [Ronald G. Ross Contact Information] http://www.brsolutions.com/email/wordpress-2.png *Blog* || < http://www.ronross.info/blog/ > http://www.RonRoss.info/blog/ http://www.brsolutions.com/email/linkedin.png *LinkedIn* || < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.linkedin.com/pub/ronald-ross/1/3b/346 http://www.brsolutions.com/email/twitter.png < https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> *Twitter* || < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross http://www.brsolutions.com/email/button-green.png *Homepage* || < http://www.ronross.info/ > http://www.RonRoss.info -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [http://www.BRSolutions.com/email/RonaldRoss.png] [http://www.brsolutions.com/email/wordpress-2.png] Blog || http://www.RonRoss.info/blog/ [http://www.brsolutions.com/email/linkedin.png] LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 [http://www.brsolutions.com/email/twitter.png] Twitter || https://twitter.com/Ronald_G_Ross [http://www.brsolutions.com/email/button-green.png] Homepage || http://www.RonRoss.info X-Yahoo-Newman-Id: 262228.16611.bm@omp1017.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316589225; bh=24C6eyuwNgkw9mbDos1g2HV3PPb5qTPIZDC6v7/8KsY=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:References:In-Reply-To:Mime-Version:Content-Type:Message-Id:Content-Transfer-Encoding:Cc:X-Mailer:From:Subject:Date:To; b=2TDl4xNI2gfxWqcxF0nCKopXoOSC7JkqMnXO6AwIcgLrbO/E4nK6wunUVBsb+PtDFOdQvKYu3kpY6pWc7pBQVbV6BFDcLAdpUAFxefRWhKshq9rIrt7nmWy2b0hyvHLonmQQaLvHIcEaa/tvm+ZzhBWB4LIc9GJ9OxNA6qA+bqI= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: YT0BaZEVM1kgJRcM3FG9QuPUDFfiSjPCM_bHgEhSAVr64Gx ufm5kwoBvjzqmIJWQPdb8MZ69WW9XNpKamBenov2PYTuaDd7hcVTjQrfNrvb xtiXHPWYrZyACl4UjfsQa3Ov9ZDY0q0sXfoitXQjPnc4BJ4qzUnW8MgoO7uA q6h_0Kxu7MHmtfMBDmcbi40C5yHiz4pxG4BdRB6YH7M5BAH9IC5aKWp_hnhD d_yfluDcHgRmzpqdNWqml3RhuKKmwa043ttszCbhY_pZBpC_jg9XVZLUx9y5 u48PKQ77HmWkx34GU8ebaTm48lSxCtG5b_QOWm6D0uwpS9ky7qgZf9nbe2bB v8lUYF1OTBa91EnxB38XAZwp0c9rQiMr.Xk0_60v0BhM6SKIajSvF.x8rFUs 3CDZmeN_dMOgNyVxgSQYlXTFvV2Q- X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI Cc: "sbvr-rtf@omg.org" , "date-time@omg.org" X-Mailer: iPad Mail (8J2) From: Stan Hendryx Subject: Re: meaning of 'proposition' Date: Wed, 21 Sep 2011 00:17:15 -0700 To: "Barkmeyer, Edward J" X-MIME-Autoconverted: from quoted-printable to 8bit by amethyst.omg.org id p8L7DlPt029704 I'm supportive of getting a credentialed expert involved. I suspect the way to do this is to write down the model very carefully and then solicit comments from some august logicians. Meanwhile... Regarding statements that deny necessities...in the actualist view such statements cannot denote anything because a necessity is, by definition of necessity, included in each possible world, and a state and it's negation cannot coexist in the same world, by the law of the excluded middle. Including a state of affairs in a possible world necessarily precludes its negation from the possible world. Statements that deny necessities express propositions, to be sure, and can be formalized by closed logical formulations that mean the propositions. Thus we can talk and reason about such propositions, but there is no corresponding state of affairs. It is not necessary that every grammatical statement denotes a state of affairs. On the other hand, paradoxical statements are nonsense; they do not even express a proposition and cannot be formalized by any logical formulation. They also have no denotation. There are fewer constraints on language than there are on the universe, so one would expect to be able to say things that are nonsense or that are impossible. As Frege said, "It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference." What Kripke did was to expand the set of statements that have denotations to include those having denotations that are possible but do not obtain. What Plantinga did was to limit what exists to what is possible. Stan On Sep 20, 2011, at 8:35 AM, "Barkmeyer, Edward J" wrote: > Stan wrote: > > [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] > > It is a lot more that 'not exactly'. This model is importantly different from the SBVR idea that there exists a referent of the grammatically well-formed expression -- a corresponding state of affairs -- in every possible world, and the expression may even deny a necessity. We can talk about impossible situations; therefore they exist. But Terry claims to be consistent with Kripke semantics; so some part of this reading is inaccurate. > > This is why we the amateur logicians need to stop debating this stuff, find an expert or two, and lay our problem -- carefully phrased -- before them. > > -Ed > > -- > Edward J. Barkmeyer Email: edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 > Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 > ________________________________________ > From: Stan Hendryx [stan@hendryxassoc.com] > Sent: Sunday, September 18, 2011 3:49 AM > To: Ronald G. Ross > Cc: Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org > Subject: Re: meaning of 'proposition' > > Ron wrote: > Why would we want to get into 'artificial' languages ... *especially* ones that are already formally grounded?? > > Machines use different formal languages, so there is a need to translate between these languages for machines to communicate just as there is a need to translate between natural languages business people use. > >>> What about existential propositions? e.g.., John Hall exists. > > These, too, involve roles. "John Hall exists" is a statement that represents an instance of the unary fact type 'thing exists'. The role is 'thing'. The bindable target of the role in the example statement is "John Hall". The actuality denoted by the statement is John Hall. Note that SBVR does not include the fact type 'statement denotes state of affairs', so, using SBVR terms, we would say "the actuality that corresponds to the proposition expressed by the statement is John Hall." > > In his landmark paper of 1892, Frege explained that a sign has both a reference and a sense. In SBVR, I take a statement to be a sign, a state of affairs to be the reference (denotation) of the sign, and a proposition to be the sense (meaning) of the sign. I thought it might interest you to see some of Frege's own words on this (from Max Black's 1960 translation of the original German): > > "It is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which the sign refers, which may be called the reference of the sign, also what I should like to call the sense of the sign, wherein the mode of presentation is contained. In our example,...the reference of 'evening star' would be the same as that of 'morning star', but not the sense. > > "The regular connection between a sign, its sense, and its reference is of such a kind that to the sign there corresponds a definite sense and to that in turn a definite reference, while to a given reference (an object) there does not belong only a single sign. The same sense has different expressions in different languages or even in the same language...It may be granted that every grammatically well-formed expression representing a proper name always has a sense. But this is not to say that to the sense there also corresponds a reference. The words 'the celestial body most distant from the Earth' have a sense, but it is very doubtful if they also have a reference...In grasping a sense, one is not certainly assured of a reference. > [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) and says that every grammatically well-formed expression that is not paradoxical and that does not deny a necessity is certainly assured of a reference (a state of affairs) in some possible world. If the reference is included in, is part of, the actual world, the state of affairs is an actuality. This is not exactly the SBVR notion of an actuality.] > > "If words are used in the ordinary way, what one intends to speak of is their reference. It can also happen, however, that one wishes to talk about the words themselves or their sense. This happens, for instance, when the words of another are quoted..." Date: Fri, 23 Sep 2011 10:15:19 +1000 Subject: Re: meaning of 'proposition' From: Terry Halpin To: Stan Hendryx Cc: "Barkmeyer, Edward J" , "Ronald G. Ross" , sbvr-rtf@omg.org, date-time@omg.org I don't agree that the sentence 'John is in London' expresses an "atemporal proposition" that corresponds to a "primitive state of affairs". To me, the tense of a sentence conveys part of its meaning. So the present tense sentence 'John is in London' asserts that John is in London at the time the sentence was uttered. Of course, it is possible to combine tenses in a sentence, e.g. 'John is, was, or will be in London' does express a proposition that corresponds to a state of affairs. Cheers Terry On Tue, Sep 20, 2011 at 10:17 AM, Stan Hendryx wrote: The matter of "always true or false (but not both)" is explained by considering that states of affairs denoted by sentences like 'John is in London at least once a month' include the state of affairs denoted by 'John is in London', where the latter is interpreted as being the primitive notion of John being in London, regardless of time. In this context, 'is' does not mean John is in London now, just what it means for John to be in London at all. It is this primitive state of affairs that is said to occur at least once a month, but saying it recurs is just a figure of speech. Formally, such a state of affairs does not literally itself recur; no state of affairs happens more than once. Rather, the primitive state of affairs is included in each state of affairs that is a monthly occurrence. The primitive is what occurs in an occurrence. In Plantinga's theory of actualism, the primitive states of affairs are called "haecceities", which is the essential property of a thing being just what it is. 'Includes' is transitive: if A includes B and B includes C, then A includes C. Thus, the graph of the 'includes' relation in a possible world is a lattice, not a tree, because the primitives can be part of many particular occurrences. The lattice is defined by the set of states of affairs the possible world includes (the set is called the domain of the possible world) and the 'includes' relation. The possible world is the top of the lattice, so it is a sort of maximal state of affairs, being the greatest lower bound of the set under 'includes'. A possible world includes a particular state of affairs at most once, as a member of its domain. The primitive proposition 'John is in London' is true in a possible world just in case the possible world includes at least one state of affairs that includes the primitive state of affairs that John is in London. The primitive is a genuine state of affairs; it occurs for a time interval that is the convex hull of each state of affairs that includes it. Stan > -----Original Message----- > From: Barkmeyer, Edward J [mailto:edward.barkmeyer@nist.gov] > Sent: Monday, September 19, 2011 2:16 PM > To: Terry Halpin; Ronald G. Ross > Cc: Stan Hendryx; sbvr-rtf@omg.org; date-time@omg.org > Subject: RE: meaning of 'proposition' > > Terry wrote: > > > To me, a proposition is what it is that is asserted by a declarative > sentence (or the declarative component of a sentence), > > and is always true or false (but not both). The SBVR "meaning that is > true or false" seems to correspond pretty close to this. > > Well, the 'but not both' is precisely the point of the issue. In 'John is > in London at least once a month', the SBVR approach to modeling this > statement makes 'John is in London' a declarative component of a sentence, > but possibly not what Terry means by 'the declarative component of a > sentence'. If the same fact model includes 'John is not in London on > Fridays', it is clear that the component 'John is in London' cannot be > either true or false, unless it is both. In either case, it fails Terry's > proposed definition of 'proposition' and therefore is not a proposition. > It follows that it does not correspond to any 'state of affairs'. This is > the dilemma that motivated the issue. > > The particular problem here is that Terry's definition is intended to > relate the concept 'proposition' only to sentences in their entirety, i.e., > John is in London at least once a month is a proposition, but no subset of > it is, or at least, no subset is being asked whether it is. That resolves > the issue, but with the same consequence: 'John is in London' is not a > declarative sentence in the fact model, and therefore does not represent a > proposition, and thus cannot correspond to a state of affairs. So, it > doesn't support the desired SBVR semantics either. > > -Ed > > -- > Edward J. Barkmeyer Email: edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 > Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 > ________________________________________ > From: Terry Halpin [terry.halpin@logicblox.com] > Sent: Sunday, September 18, 2011 7:54 PM > To: Ronald G. Ross > Cc: Stan Hendryx; Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org > Subject: Re: meaning of 'proposition' > > Hi Ron > > < categories of proposition (relevant to SBVR) -- (1) existential > propositions, and (2) propositions involving some role? > > The current definition of "proposition:" in SBVR seems inadequate ... > "meaning thatis true or false " ... and this is one source of much > consternation. For example, that definition would seem to admit boolean > variables. I think we really need something better.>> > > I distinguish purely existential propositions (e.g. there is a country > code 'US') from propositions that predicate over individuals (e.g. Ron was > born in the USA). To me, a proposition is what it is that is asserted by a > declarative sentence (or the declarative component of a sentence), and is > always true or false (but not both). The SBVR "meaning that is true or > false" seems to correspond pretty close to this. I don't consider a > variable to be a meaning. > > Cheers > Terry > > > > On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross > > wrote: > At 05:00 PM 9/18/2011, Terry Halpin wrote: > < > Stan wrote: These, too, involve roles. "John Hall exists" is a statement > that represents an instance of the unary fact type 'thing exists'. The > role is 'thing'. The bindable target of the role in the example statement > is "John Hall". The actuality denoted by the statement is John Hall. Note > that SBVR does not include the fact type 'statement denotes state of > affairs', so, using SBVR terms, we would say "the actuality that > corresponds to the proposition expressed by the statement is John Hall." > >> > > I disagree with Stan here. I use "role" in two ways. A "fact role" is a > part played in a fact, and always involves a logical predicate (e.g. The > fact type "Person drives Car" has two roles, one being the role of driving > which is played by a person, and the other being the role of being driven > which is played by a car). A "role type" is a non-rigid type (e.g. > Student), unlike rigid types such as Person or Thing. > > The elementary fact "John Hall is alive" involves an individual and a > unary predicate "is alive", which may be treated as the role of being > alive. The existential fact "John Hall exists" involves an individual and > an existential quantifier. The usual view in logic, with which I concur, > is that existence is not a predicate (and certainly not a first-order > predicate). "Exists" is not the same kind of construct as "is alive", and > it does not involve a role. If we use the individual constant "j" for John > Hall, then "John Hall is alive" may be formalized as "j isAlive" (in > postfix notation), but "John Hall exists" is formalized as "Exists x x = > j". > > Terry, > > Would it be fair to say then that there are two categories of > proposition (relevant to SBVR) -- (1) existential propositions, and (2) > propositions involving some role? > > The current definition of "proposition:" in SBVR seems > inadequate ... "meaning that is true or false " ... and this is one source > of much consternation. For example, that definition would seem to admit > boolean variables. I think we really need something better. > > Thanks, > Ron > > > Cheers > Terry > > On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx > > wrote: > > Ron wrote: > Why would we want to get into 'artificial' languages ... *especially* ones > that are already formally grounded?? > > Machines use different formal languages, so there is a need to translate > between these languages for machines to communicate just as there is a > need to translate between natural languages business people use. > > >>What about existential propositions? e.g.., John Hall exists. > > These, too, involve roles. "John Hall exists" is a statement that > represents an instance of the unary fact type 'thing exists'. The role is > 'thing'. The bindable target of the role in the example statement is "John > Hall". The actuality denoted by the statement is John Hall. Note that SBVR > does not include the fact type 'statement denotes state of affairs', so, > using SBVR terms, we would say "the actuality that corresponds to the > proposition expressed by the statement is John Hall." > > In his landmark paper of 1892, Frege explained that a sign has both a > reference and a sense. In SBVR, I take a statement to be a sign, a state > of affairs to be the reference (denotation) of the sign, and a proposition > to be the sense (meaning) of the sign. I thought it might interest you to > see some of Frege's own words on this (from Max Black's 1960 translation > of the original German): > > "It is natural, now, to think of there being connected with a sign (name, > combination of words, letter), besides that to which the sign refers, > which may be called the reference of the sign, also what I should like to > call the sense of the sign, wherein the mode of presentation is contained. > In our example,...the reference of 'evening star' would be the same as > that of 'morning star', but not the sense. > > "The regular connection between a sign, its sense, and its reference is of > such a kind that to the sign there corresponds a definite sense and to > that in turn a definite reference, while to a given reference (an object) > there does not belong only a single sign. The same sense has different > expressions in different languages or even in the same language...It may > be granted that every grammatically well-formed expression representing a > proper name always has a sense. But this is not to say that to the sense > there also corresponds a reference. The words 'the celestial body most > distant from the Earth' have a sense, but it is very doubtful if they also > have a reference...In grasping a sense, one is not certainly assured of a > reference. > [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) > and says that every grammatically well-formed expression that is not > paradoxical and that does not deny a necessity is certainly assured of a > reference (a state of affairs) in some possible world. If the reference is > included in, is part of, the actual world, the state of affairs is an > actuality. This is not exactly the SBVR notion of an actuality.] > > "If words are used in the ordinary way, what one intends to speak of is > their reference. It can also happen, however, that one wishes to talk > about the words themselves or their sense. This happens, for instance, > when the words of another are quoted..." > --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in > Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), > pp. 25-50 > > Thanks, Ron, for the clarification about the context you mentioned. I > understand from your comment that the idea of a proposition giving truth > conditions is new in the SBVR discussions. If a fact model meets the truth > conditions, the proposition is true for that fact model, for the possible > world modeled by the fact model. What it means for a fact model to meet > the conditions of a proposition is spelled out in the model-theoretic > semantics of SBVR, which, as Ed pointed out, is not well documented. This > is an area for future work on SBVR. The model theory of a language is a > significant part of it's formal grounding, important to interpreting the > language and getting translations correct. Not all 'artificial' languages > have a model theory, e.g. UML does not. > > Stan > > On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" > > wrote: > > >>Good discussion. Some notes inserted like this below. > > Ron > > At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: > Stan Hendryx wrote: > Ed wrote: > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, > > Three SBVR fact types are particularly relevant here: > statement expresses proposition > closed logical formulation formalizes statement > closed logical formulation means proposition > > Only in passing. The proposition (meaning) is the same, whether it is > expressed in English or Structured English or Chinese, or SBVR clause 9 > XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR > is that it doesn't seem to have considered that all of these are > Statements -- expressions representing propositions -- and several of them > are at least as formally grounded as SBVR aspires to be. > > These are related by this rule, which is a composite of several SBVR > necessities: > Each statement expresses exactly one proposition and the statement is > formalized by a closed logical formulation that means just that > proposition. > > A statement in CLIF doesn't need to be formalized by an SBVR closed > logical formulation. It is formalized by the language. Statement means > 'representation of a proposition' -- any representation. > > This should be qualified by saying "each statement that is not paradoxical > expresses...", since paradoxes are nonsensical, not having a meaning, no > logical formulation, e.g. "this sentence is false." Such sentences do not > express propositions, either. > > A closed logical formulation is a representation, a translation, of the > statement it formalizes into the language of predicate logic. > > Well, no. It formalizes it into a language defined in clause 9, 13 and 15 > whose semantics is said to be a dialect of predicate logic. There is no > 'the language of predicate logic'; there are many. And whether the logic > language that Clause 9 introduces has a well-defined formal semantics is > still debatable. None has ever been formally written down. > > A CLIF representation of a proposition is a representation in a standard > language for predicate logic. And you yourself used Quine's mathematical > notation to state propositions. Surely that didn't need to be formalized > in SBVR LRMV to be formalized in the language of predicate logic. > > This is all about thinking that Statement means 'natural language > statement' or perhaps 'Structured English statement', and that idea > pervades SBVR. In Date/Time, we have expressed most of the formal > propositions in SBVR SE, OCL and CLIF, of which the last two are formal > and have formal semantics. In point of fact, we have no tool that > generates the SBVR LRMV closed logical formulations. But hope springs > eternal, at least outside of NIST. > > Ron wrote: > "Truth conditions" is new to me in this context. > > This idea came from G. Frege ca. 1892. His work, which includes the > invention of predicate calculus, is the foundation most subsequent work in > logic and formal semantics. > > >>Next time when I say "this context" I will take the time to be more > clear for you (Stan). "This context" means the SBVR-related e-mails I have > been recently reading regarding the definition of "proposition" in SBVR. > There, better? > > > > Certainly logical formulations should be considered out of bounds for > defining what a proposition is. > > On the contrary, a closed logical formulation that means a proposition > definitively says what the proposition is. SBVR gives a definition of the > concept 'proposition': meaning that is true or false. > > I agree with Ron. The point of all this is that the issue of what a > proposition is does not depend on how it is represented. > > >>Thank you, Ed. That's so obvious it should speak for itself (but of > course it can't). > > > And it certainly doesn't depend on its being represented in an XML form > which no tools generate. 'John is in London' definitively says what the > proposition is, at least to speakers of English. You cannot express it > any more clearly in predicate logic, because that would also require that > you explain that the symbol 'John' is an individual concept (logical > constant) that refers to a specific person, and similarly 'London', and > that the predicate symbol 'lives in' refers to the verb concept that is > designated 'lives in' in English. The formal language is just a grammar > that eliminates certain ambiguities in the interpretation, assuming that > you know what the terminal symbols mean. > > To understand the relationship between fact types and their roles, and > propositions, consider that an expression of an instance of a fact type is > a sentence, an atomic sentence. For example, the sentence "Stan was born > in the US," expresses an instance of the binary fact type 'person was born > in country'. Such a sentence expresses a proposition. The sentence is > formalized by an atomic formulation with role bindings, i.e. it is a > closed logical formulation; I call it a closed atomic formulation (my term, > not SBVR). Each role binding is of a role of the fact type. Each role > binding binds to a bindable target, which is either a bound variable, an > individual (Stan and the US in the example) or an identifying expression. > The closed atomic formulation means the proposition that is expressed by > the sentence. The proposition corresponds to a state of affairs. The state > of affairs is the instance of the fact type. If the state of affairs is an > actuality, the proposition is true. To sum up, an instance of a fact type > is a state of affairs that corresponds to the proposition that is > expressed by the statement of the instance of the fact type. > > I thought SBVR was clearer on this. An instance of a fact type is an > actuality that is understood as involving some things in the roles > specified in the fact type form(s). > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? > > If the statement of the proposition is formal in the SBVR sense (colored), > then the proposition will always involve some role, > > I agree. "Murphy's Law" is an expression that denotes a proposition, and > that exrpression does not involve any thing in any role. But the meaning > that is "Murphy's Law" depends on an intransitive verb concept and > involves the general pronoun 'anything' in the sole role in the unary fact > type 'thing goes wrong'. That is one more reason to stop worrying about > the nature of the expression. > The nature of the proposition as a meaning is to conceptualize a situation > as involving things in roles. I know of no counterexamples. > >>What about existential propositions? e.g.., John Hall exists. ... Some > person exists. > > From a business (not logician's) point of view, these to me to involve > roles too. So I agree ... I can think of no counterexamples. Can anyone > else? > > If not, I believe this is a fundamental characteristic in the meaning of > 'proposition'. It seems right at the heart of what SBVR is about. > > > -Ed > > since every closed logical formulation involves at least one fact type > (verb concept) and it's atomic formulation. We know this because every > statement has a verb and by hypothesis, the verb is formalized. In SBVR, > verbs are formalized as fact type designations, and each fact type has at > least one role. > Stan > > On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > < > mailto:rross@brsolutions.com>> wrote: > > At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: > > > Stan Hendryx wrote: > Keri wrote: > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > Yes. The meaning of a proposition is its truth conditions, what will make > it true or false. These conditions are stated precisely by the closed > logical formulation that formalizes a statement that expresses the > proposition. Truth conditions are independent of any facts. > > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, Stan's position is the basis > for the issue. The nature of the proposition is not that it has a truth > value but rather that it is a set of conditions that determines a truth > value in a given world. The issue as written described that set of > conditions as the 'conceptualization of a situation', but perhaps Stan's > characterization is a bit clearer. Those conditions involve things > playing roles (which, according to SBVR is the nature of a state of > affairs). > The truth value of a proposition, it being true or false, depends on the > facts in each fact model in which it is evaluated. It may be true in some > fact models and false in others. It may be unknown. We know what a > proposition means even if it is false or even if its truth value is > unknown. > I hope this helps clarify what Ron meant. > > Stan is probably right that I misunderstood Ron's point. Ron apparently > meant that the meaning that is a proposition is independent of its truth > value. > > Yes, I meant: By definition, a meaning that is a proposition is > independent of its truth value. > > We fully agree. Does Ron agree with Stan that the nature of the meaning > is the 'truth conditions'? And if we agree, can we write a resolution to > 16526 to say that? > > "Truth conditions" is new to me in this context. Certainly logical > formulations should be considered out of bounds for defining what a > proposition is. Ed, you raised some interesting points in your earlier e- > mail response about roles and fact types with respect to defining > propositions I want to hear discussion about. > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? I say > cautiously because (a) I don't know where that takes things, and (b) I > haven't checked to see if the definitions would become circular. > > Ron > > > -Ed > > > Stan > > > > On Sep 16, 2011, at 10:59 AM, keri > < mailto:keri_ah@mac.com> < < > mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: > > Mark, > > Here are some of the points I captured in my notes. (I don't show that > the bits about "independent of any Universe of Discourse" was part of what > Ron said, but perhaps so and I missed that. From the discussion around > the example proposition statements it did appear that they pertained to > some world/UOD.) > > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > I also captured points such as: > * Being true or false does not depend on (a) if it's known that > it's true or false or (b) that it's interesting that it's true > or false. > * Being true or false is different from ambiguity. > > In regards to the example "John is in London." I have to conclude that the > reason it is a proposition (without knowing whether or not John is > actually in London) is because this statement (of proposition) is /based > on/ the vocabulary's fact type "person is in place" and the further aspect > that a person must be in exactly one place. If this is the meaning that > is understood then it can be seen that any statement about some person > being in some place must be either true or false, without knowing which it > is. (I think that's how it works.) When we exchange examples > propositions we have to have the meaning come from the vocabulary concepts > that the proposition is /based on/. Otherwise, there is no meaning. > > - Keri > > > On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: > > During today's RTF discussion, we discussed this example from Ed's note of > September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My > purpose in writing this email is to summarize what I think I heard. > Please comment if you think I got it wrong. > > > Given a fact model comprising: John is in London during 5-8 September > > 2011. John is in St. Imier during 9-10 September 2011. The time of > > interest in the Universe of Discourse is Sept 5-10, 2011. The > > proposition expressed by 'John is in London' is used in representing > > this fact model, both in SBVR examples, and formally in Date/Time. Is > > it true or false? I believe that in that actual world, the proposition > > is neither true nor false. Therefore, it is a counterexample to what > > Don says. > > 1. SBVR's definition of "proposition" is "meaning that is true or false". > Despite the use of the word "is" in the definition, this definition does > not mean that you have to interpret an expression against a Universe of > Discourse to determine whether the expression expresses a proposition. > Ron Ross said something to the effect that it is very important that a > proposition is one independent of any Universe of Discourse. I think he's > right -- but the current definition is not clear about that. I understand > that the RTF will propose a change to clarify this point. > > 2. Regarding the example given above, what I heard is that whether "John > is in London" may not be known in the model, and is not inferable solely > from the two facts given. So the expression may not be decidable, but it > is still expresses a proposition. > > > Is this a correct summary? > -------------------------------- > Mark H. Linehan > STSM, Model Driven Business Transformation > IBM Research > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov < mailto:edbark@nist.gov> > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > [Ronald G. Ross Contact Information] > http://www.brsolutions.com/email/wordpress-2.png *Blog* || > < http://www.ronross.info/blog/ > > http://www.RonRoss.info/blog/ > http://www.brsolutions.com/email/linkedin.png *LinkedIn* || > < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.brsolutions.com/email/twitter.png < > https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> > *Twitter* || > < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross > http://www.brsolutions.com/email/button-green.png *Homepage* > || < http://www.ronross.info/ > > http://www.RonRoss.info > > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 > Twitter || https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > Twitter || > https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > X-Yahoo-Newman-Id: 620168.82688.bm@omp1024.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316740112; bh=By5cqOWmZ5Dhiq1A7Ut3PpbknvJOzluuNfRm8NVq3mc=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:From:To:Cc:References:Subject:Date:Organization:Message-ID:MIME-Version:Content-Type:X-Mailer:X-MimeOLE:Thread-Index:In-Reply-To; b=6gjeX0U4i+vqgqdiuGIDRIOMpCw3BGb9YkR6qDsghqhJAhAYBknBcwxMbGG3OUPZFN6rFSTtJF/M+89N3y/ffZ+r26EXuaRmE4TuxFvyQUOeEbpinNm8j1Vzccg7/X1F/RqEFz+EPTWdGUGv6zNx9VWm4RhB+jMSUAHL+QZ0xjc= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: cRkw7aYVM1mtr.q6x5DtAMVPfwnh6oCfzDfdrSGWwwkVx7X 7ip_zEHfwePDHrbz35oOa0YUGmu0KhUVMAZrElWUZXFflm8kZ0bLxS94wseN ojjWisgeGMrtUZR3jKofUzIzYTQSngVFz4scQyo7nDFIyLK5HPj3TcXIVVxk h1wC.wIRlcUzh_4bZe2ST.7Sx_Euc1zXYVwDWyOpvH7FKFhSArXwXRdh0Ur3 .ik5Ijjn6i.fYwybpZMZhPPOB6DOZ_m5ptCFa5LEOYt8CBgDphhp4J6yjh7H pBRihxqPN9.F7l7tDOV0FqcWaijuAaX2YXL_dIzX6pMpZP4hQTN9kWIpVgUJ OTKM15jgVS3OAi16.Upw0N5sxd9tOeQBCyjElD1sXEiY8aAZOxyiADWQUwad VBK3Hz2KpkSr0jvRQTzgp4x56Wg-- X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI From: "Stan Hendryx" To: "'Terry Halpin'" Cc: "'Barkmeyer, Edward J'" , "'Ronald G. Ross'" , , Subject: RE: meaning of 'proposition' Date: Thu, 22 Sep 2011 18:08:19 -0700 Organization: Hendryx & Associates X-Mailer: Microsoft Office Outlook 11 Thread-Index: Acx5heLTUpv0c2/1RMy0p+nfQB5XfgAAQDfQ Terry, To me, too, the tense ordinarily conveys part of the meaning. What I was talking about as the primitive notion that is included is a tenseless concept of, say, John being in London. This primitive notion corresponds to Plantinga.s .haecceity.. Thus, .John is in London. (meaning John is in London now) includes the primitive state of affairs (that might be better stated in English as) .John being in London.. I tried overloading .is. to express the no-tense idea of a haecceity, but that didn.t work; it is too confusing. I.ll switch to using the gerund form to express a haecceity, which seems to be timeless. Plantinga never said how to express a haecceity in English, so I.m making it up. I.ll try again using a gerund: Every proposition that is about John being in London, whenever, corresponds to a state of affairs that includes the state of affairs .John being in London.. Thus, the SOA .John is in London. includes the SOA .John being in London.. So does .John was in London last Thursday.. I formulate .John is in London. as .The time interval of John being in London overlaps the current time.; or ....overlaps last Thursday.; or ....overlaps .. If John is in London often, then the time interval of John being in London overlaps the time interval of each visit, i.e. is the convex hull of the time intervals of all of the visits. We say that .John being in London. .recurs., but of course it doesn.t recur in the literal sense of the word -- no state of affairs literally happens more than once; for a state of affairs to recur means the state of affairs is part of more than one other state of affairs and its time interval overlaps that of each other state of affairs of which it is part. (.SOA2 is part of SOA1. is a synonymous form of .SOA1 includes SOA2., a reverse reading.) I hope this helps clarify what I meant. Best regards, Stan -------------------------------------------------------------------------------- From: Terry Halpin [mailto:terry.halpin@logicblox.com] Sent: Thursday, September 22, 2011 5:15 PM To: Stan Hendryx Cc: Barkmeyer, Edward J; Ronald G. Ross; sbvr-rtf@omg.org; date-time@omg.org Subject: Re: meaning of 'proposition' I don't agree that the sentence 'John is in London' expresses an "atemporal proposition" that corresponds to a "primitive state of affairs". To me, the tense of a sentence conveys part of its meaning. So the present tense sentence 'John is in London' asserts that John is in London at the time the sentence was uttered. Of course, it is possible to combine tenses in a sentence, e.g. 'John is, was, or will be in London' does express a proposition that corresponds to a state of affairs. Cheers Terry On Tue, Sep 20, 2011 at 10:17 AM, Stan Hendryx wrote: The matter of "always true or false (but not both)" is explained by considering that states of affairs denoted by sentences like 'John is in London at least once a month' include the state of affairs denoted by 'John is in London', where the latter is interpreted as being the primitive notion of John being in London, regardless of time. In this context, 'is' does not mean John is in London now, just what it means for John to be in London at all. It is this primitive state of affairs that is said to occur at least once a month, but saying it recurs is just a figure of speech. Formally, such a state of affairs does not literally itself recur; no state of affairs happens more than once. Rather, the primitive state of affairs is included in each state of affairs that is a monthly occurrence. The primitive is what occurs in an occurrence. In Plantinga's theory of actualism, the primitive states of affairs are called "haecceities", which is the essential property of a thing being just what it is. 'Includes' is transitive: if A includes B and B includes C, then A includes C. Thus, the graph of the 'includes' relation in a possible world is a lattice, not a tree, because the primitives can be part of many particular occurrences. The lattice is defined by the set of states of affairs the possible world includes (the set is called the domain of the possible world) and the 'includes' relation. The possible world is the top of the lattice, so it is a sort of maximal state of affairs, being the greatest lower bound of the set under 'includes'. A possible world includes a particular state of affairs at most once, as a member of its domain. The primitive proposition 'John is in London' is true in a possible world just in case the possible world includes at least one state of affairs that includes the primitive state of affairs that John is in London. The primitive is a genuine state of affairs; it occurs for a time interval that is the convex hull of each state of affairs that includes it. Stan > -----Original Message----- > From: Barkmeyer, Edward J [mailto:edward.barkmeyer@nist.gov] > Sent: Monday, September 19, 2011 2:16 PM > To: Terry Halpin; Ronald G. Ross > Cc: Stan Hendryx; sbvr-rtf@omg.org; date-time@omg.org > Subject: RE: meaning of 'proposition' > > Terry wrote: > > > To me, a proposition is what it is that is asserted by a declarative > sentence (or the declarative component of a sentence), > > and is always true or false (but not both). The SBVR "meaning that is > true or false" seems to correspond pretty close to this. > > Well, the 'but not both' is precisely the point of the issue. In 'John is > in London at least once a month', the SBVR approach to modeling this > statement makes 'John is in London' a declarative component of a sentence, > but possibly not what Terry means by 'the declarative component of a > sentence'. If the same fact model includes 'John is not in London on > Fridays', it is clear that the component 'John is in London' cannot be > either true or false, unless it is both. In either case, it fails Terry's > proposed definition of 'proposition' and therefore is not a proposition. > It follows that it does not correspond to any 'state of affairs'. This is > the dilemma that motivated the issue. > > The particular problem here is that Terry's definition is intended to > relate the concept 'proposition' only to sentences in their entirety, i.e., > John is in London at least once a month is a proposition, but no subset of > it is, or at least, no subset is being asked whether it is. That resolves > the issue, but with the same consequence: 'John is in London' is not a > declarative sentence in the fact model, and therefore does not represent a > proposition, and thus cannot correspond to a state of affairs. So, it > doesn't support the desired SBVR semantics either. > > -Ed > > -- > Edward J. Barkmeyer Email: edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 > Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 > ________________________________________ > From: Terry Halpin [terry.halpin@logicblox.com] > Sent: Sunday, September 18, 2011 7:54 PM > To: Ronald G. Ross > Cc: Stan Hendryx; Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org > Subject: Re: meaning of 'proposition' > > Hi Ron > > < categories of proposition (relevant to SBVR) -- (1) existential > propositions, and (2) propositions involving some role? > > The current definition of "proposition:" in SBVR seems inadequate ... > "meaning thatis true or false " ... and this is one source of much > consternation. For example, that definition would seem to admit boolean > variables. I think we really need something better.>> > > I distinguish purely existential propositions (e.g. there is a country > code 'US') from propositions that predicate over individuals (e.g. Ron was > born in the USA). To me, a proposition is what it is that is asserted by a > declarative sentence (or the declarative component of a sentence), and is > always true or false (but not both). The SBVR "meaning that is true or > false" seems to correspond pretty close to this. I don't consider a > variable to be a meaning. > > Cheers > Terry > > > > On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross > > wrote: > At 05:00 PM 9/18/2011, Terry Halpin wrote: > < > Stan wrote: These, too, involve roles. "John Hall exists" is a statement > that represents an instance of the unary fact type 'thing exists'. The > role is 'thing'. The bindable target of the role in the example statement > is "John Hall". The actuality denoted by the statement is John Hall. Note > that SBVR does not include the fact type 'statement denotes state of > affairs', so, using SBVR terms, we would say "the actuality that > corresponds to the proposition expressed by the statement is John Hall." > >> > > I disagree with Stan here. I use "role" in two ways. A "fact role" is a > part played in a fact, and always involves a logical predicate (e.g. The > fact type "Person drives Car" has two roles, one being the role of driving > which is played by a person, and the other being the role of being driven > which is played by a car). A "role type" is a non-rigid type (e.g. > Student), unlike rigid types such as Person or Thing. > > The elementary fact "John Hall is alive" involves an individual and a > unary predicate "is alive", which may be treated as the role of being > alive. The existential fact "John Hall exists" involves an individual and > an existential quantifier. The usual view in logic, with which I concur, > is that existence is not a predicate (and certainly not a first-order > predicate). "Exists" is not the same kind of construct as "is alive", and > it does not involve a role. If we use the individual constant "j" for John > Hall, then "John Hall is alive" may be formalized as "j isAlive" (in > postfix notation), but "John Hall exists" is formalized as "Exists x x = > j". > > Terry, > > Would it be fair to say then that there are two categories of > proposition (relevant to SBVR) -- (1) existential propositions, and (2) > propositions involving some role? > > The current definition of "proposition:" in SBVR seems > inadequate ... "meaning that is true or false " ... and this is one source > of much consternation. For example, that definition would seem to admit > boolean variables. I think we really need something better. > > Thanks, > Ron > > > Cheers > Terry > > On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx > > wrote: > > Ron wrote: > Why would we want to get into 'artificial' languages ... *especially* ones > that are already formally grounded?? > > Machines use different formal languages, so there is a need to translate > between these languages for machines to communicate just as there is a > need to translate between natural languages business people use. > > >>What about existential propositions? e.g.., John Hall exists. > > These, too, involve roles. "John Hall exists" is a statement that > represents an instance of the unary fact type 'thing exists'. The role is > 'thing'. The bindable target of the role in the example statement is "John > Hall". The actuality denoted by the statement is John Hall. Note that SBVR > does not include the fact type 'statement denotes state of affairs', so, > using SBVR terms, we would say "the actuality that corresponds to the > proposition expressed by the statement is John Hall." > > In his landmark paper of 1892, Frege explained that a sign has both a > reference and a sense. In SBVR, I take a statement to be a sign, a state > of affairs to be the reference (denotation) of the sign, and a proposition > to be the sense (meaning) of the sign. I thought it might interest you to > see some of Frege's own words on this (from Max Black's 1960 translation > of the original German): > > "It is natural, now, to think of there being connected with a sign (name, > combination of words, letter), besides that to which the sign refers, > which may be called the reference of the sign, also what I should like to > call the sense of the sign, wherein the mode of presentation is contained. > In our example,...the reference of 'evening star' would be the same as > that of 'morning star', but not the sense. > > "The regular connection between a sign, its sense, and its reference is of > such a kind that to the sign there corresponds a definite sense and to > that in turn a definite reference, while to a given reference (an object) > there does not belong only a single sign. The same sense has different > expressions in different languages or even in the same language...It may > be granted that every grammatically well-formed expression representing a > proper name always has a sense. But this is not to say that to the sense > there also corresponds a reference. The words 'the celestial body most > distant from the Earth' have a sense, but it is very doubtful if they also > have a reference...In grasping a sense, one is not certainly assured of a > reference. > [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) > and says that every grammatically well-formed expression that is not > paradoxical and that does not deny a necessity is certainly assured of a > reference (a state of affairs) in some possible world. If the reference is > included in, is part of, the actual world, the state of affairs is an > actuality. This is not exactly the SBVR notion of an actuality.] > > "If words are used in the ordinary way, what one intends to speak of is > their reference. It can also happen, however, that one wishes to talk > about the words themselves or their sense. This happens, for instance, > when the words of another are quoted..." > --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in > Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), > pp. 25-50 > > Thanks, Ron, for the clarification about the context you mentioned. I > understand from your comment that the idea of a proposition giving truth > conditions is new in the SBVR discussions. If a fact model meets the truth > conditions, the proposition is true for that fact model, for the possible > world modeled by the fact model. What it means for a fact model to meet > the conditions of a proposition is spelled out in the model-theoretic > semantics of SBVR, which, as Ed pointed out, is not well documented. This > is an area for future work on SBVR. The model theory of a language is a > significant part of it's formal grounding, important to interpreting the > language and getting translations correct. Not all 'artificial' languages > have a model theory, e.g. UML does not. > > Stan > > On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" > > wrote: > > >>Good discussion. Some notes inserted like this below. > > Ron > > At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: > Stan Hendryx wrote: > Ed wrote: > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, > > Three SBVR fact types are particularly relevant here: > statement expresses proposition > closed logical formulation formalizes statement > closed logical formulation means proposition > > Only in passing. The proposition (meaning) is the same, whether it is > expressed in English or Structured English or Chinese, or SBVR clause 9 > XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR > is that it doesn't seem to have considered that all of these are > Statements -- expressions representing propositions -- and several of them > are at least as formally grounded as SBVR aspires to be. > > These are related by this rule, which is a composite of several SBVR > necessities: > Each statement expresses exactly one proposition and the statement is > formalized by a closed logical formulation that means just that > proposition. > > A statement in CLIF doesn't need to be formalized by an SBVR closed > logical formulation. It is formalized by the language. Statement means > 'representation of a proposition' -- any representation. > > This should be qualified by saying "each statement that is not paradoxical > expresses...", since paradoxes are nonsensical, not having a meaning, no > logical formulation, e.g. "this sentence is false." Such sentences do not > express propositions, either. > > A closed logical formulation is a representation, a translation, of the > statement it formalizes into the language of predicate logic. > > Well, no. It formalizes it into a language defined in clause 9, 13 and 15 > whose semantics is said to be a dialect of predicate logic. There is no > 'the language of predicate logic'; there are many. And whether the logic > language that Clause 9 introduces has a well-defined formal semantics is > still debatable. None has ever been formally written down. > > A CLIF representation of a proposition is a representation in a standard > language for predicate logic. And you yourself used Quine's mathematical > notation to state propositions. Surely that didn't need to be formalized > in SBVR LRMV to be formalized in the language of predicate logic. > > This is all about thinking that Statement means 'natural language > statement' or perhaps 'Structured English statement', and that idea > pervades SBVR. In Date/Time, we have expressed most of the formal > propositions in SBVR SE, OCL and CLIF, of which the last two are formal > and have formal semantics. In point of fact, we have no tool that > generates the SBVR LRMV closed logical formulations. But hope springs > eternal, at least outside of NIST. > > Ron wrote: > "Truth conditions" is new to me in this context. > > This idea came from G. Frege ca. 1892. His work, which includes the > invention of predicate calculus, is the foundation most subsequent work in > logic and formal semantics. > > >>Next time when I say "this context" I will take the time to be more > clear for you (Stan). "This context" means the SBVR-related e-mails I have > been recently reading regarding the definition of "proposition" in SBVR. > There, better? > > > > Certainly logical formulations should be considered out of bounds for > defining what a proposition is. > > On the contrary, a closed logical formulation that means a proposition > definitively says what the proposition is. SBVR gives a definition of the > concept 'proposition': meaning that is true or false. > > I agree with Ron. The point of all this is that the issue of what a > proposition is does not depend on how it is represented. > > >>Thank you, Ed. That's so obvious it should speak for itself (but of > course it can't). > > > And it certainly doesn't depend on its being represented in an XML form > which no tools generate. 'John is in London' definitively says what the > proposition is, at least to speakers of English. You cannot express it > any more clearly in predicate logic, because that would also require that > you explain that the symbol 'John' is an individual concept (logical > constant) that refers to a specific person, and similarly 'London', and > that the predicate symbol 'lives in' refers to the verb concept that is > designated 'lives in' in English. The formal language is just a grammar > that eliminates certain ambiguities in the interpretation, assuming that > you know what the terminal symbols mean. > > To understand the relationship between fact types and their roles, and > propositions, consider that an expression of an instance of a fact type is > a sentence, an atomic sentence. For example, the sentence "Stan was born > in the US," expresses an instance of the binary fact type 'person was born > in country'. Such a sentence expresses a proposition. The sentence is > formalized by an atomic formulation with role bindings, i.e. it is a > closed logical formulation; I call it a closed atomic formulation (my term, > not SBVR). Each role binding is of a role of the fact type. Each role > binding binds to a bindable target, which is either a bound variable, an > individual (Stan and the US in the example) or an identifying expression. > The closed atomic formulation means the proposition that is expressed by > the sentence. The proposition corresponds to a state of affairs. The state > of affairs is the instance of the fact type. If the state of affairs is an > actuality, the proposition is true. To sum up, an instance of a fact type > is a state of affairs that corresponds to the proposition that is > expressed by the statement of the instance of the fact type. > > I thought SBVR was clearer on this. An instance of a fact type is an > actuality that is understood as involving some things in the roles > specified in the fact type form(s). > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? > > If the statement of the proposition is formal in the SBVR sense (colored), > then the proposition will always involve some role, > > I agree. "Murphy's Law" is an expression that denotes a proposition, and > that exrpression does not involve any thing in any role. But the meaning > that is "Murphy's Law" depends on an intransitive verb concept and > involves the general pronoun 'anything' in the sole role in the unary fact > type 'thing goes wrong'. That is one more reason to stop worrying about > the nature of the expression. > The nature of the proposition as a meaning is to conceptualize a situation > as involving things in roles. I know of no counterexamples. > >>What about existential propositions? e.g.., John Hall exists. ... Some > person exists. > > From a business (not logician's) point of view, these to me to involve > roles too. So I agree ... I can think of no counterexamples. Can anyone > else? > > If not, I believe this is a fundamental characteristic in the meaning of > 'proposition'. It seems right at the heart of what SBVR is about. > > > -Ed > > since every closed logical formulation involves at least one fact type > (verb concept) and it's atomic formulation. We know this because every > statement has a verb and by hypothesis, the verb is formalized. In SBVR, > verbs are formalized as fact type designations, and each fact type has at > least one role. > Stan > > On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > < > mailto:rross@brsolutions.com>> wrote: > > At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: > > > Stan Hendryx wrote: > Keri wrote: > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > Yes. The meaning of a proposition is its truth conditions, what will make > it true or false. These conditions are stated precisely by the closed > logical formulation that formalizes a statement that expresses the > proposition. Truth conditions are independent of any facts. > > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, Stan's position is the basis > for the issue. The nature of the proposition is not that it has a truth > value but rather that it is a set of conditions that determines a truth > value in a given world. The issue as written described that set of > conditions as the 'conceptualization of a situation', but perhaps Stan's > characterization is a bit clearer. Those conditions involve things > playing roles (which, according to SBVR is the nature of a state of > affairs). > The truth value of a proposition, it being true or false, depends on the > facts in each fact model in which it is evaluated. It may be true in some > fact models and false in others. It may be unknown. We know what a > proposition means even if it is false or even if its truth value is > unknown. > I hope this helps clarify what Ron meant. > > Stan is probably right that I misunderstood Ron's point. Ron apparently > meant that the meaning that is a proposition is independent of its truth > value. > > Yes, I meant: By definition, a meaning that is a proposition is > independent of its truth value. > > We fully agree. Does Ron agree with Stan that the nature of the meaning > is the 'truth conditions'? And if we agree, can we write a resolution to > 16526 to say that? > > "Truth conditions" is new to me in this context. Certainly logical > formulations should be considered out of bounds for defining what a > proposition is. Ed, you raised some interesting points in your earlier e- > mail response about roles and fact types with respect to defining > propositions I want to hear discussion about. > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? I say > cautiously because (a) I don't know where that takes things, and (b) I > haven't checked to see if the definitions would become circular. > > Ron > > > -Ed > > > Stan > > > > On Sep 16, 2011, at 10:59 AM, keri > < mailto:keri_ah@mac.com> < < > mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: > > Mark, > > Here are some of the points I captured in my notes. (I don't show that > the bits about "independent of any Universe of Discourse" was part of what > Ron said, but perhaps so and I missed that. From the discussion around > the example proposition statements it did appear that they pertained to > some world/UOD.) > > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > I also captured points such as: > * Being true or false does not depend on (a) if it's known that > it's true or false or (b) that it's interesting that it's true > or false. > * Being true or false is different from ambiguity. > > In regards to the example "John is in London." I have to conclude that the > reason it is a proposition (without knowing whether or not John is > actually in London) is because this statement (of proposition) is /based > on/ the vocabulary's fact type "person is in place" and the further aspect > that a person must be in exactly one place. If this is the meaning that > is understood then it can be seen that any statement about some person > being in some place must be either true or false, without knowing which it > is. (I think that's how it works.) When we exchange examples > propositions we have to have the meaning come from the vocabulary concepts > that the proposition is /based on/. Otherwise, there is no meaning. > > - Keri > > > On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: > > During today's RTF discussion, we discussed this example from Ed's note of > September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My > purpose in writing this email is to summarize what I think I heard. > Please comment if you think I got it wrong. > > > Given a fact model comprising: John is in London during 5-8 September > > 2011. John is in St. Imier during 9-10 September 2011. The time of > > interest in the Universe of Discourse is Sept 5-10, 2011. The > > proposition expressed by 'John is in London' is used in representing > > this fact model, both in SBVR examples, and formally in Date/Time. Is > > it true or false? I believe that in that actual world, the proposition > > is neither true nor false. Therefore, it is a counterexample to what > > Don says. > > 1. SBVR's definition of "proposition" is "meaning that is true or false". > Despite the use of the word "is" in the definition, this definition does > not mean that you have to interpret an expression against a Universe of > Discourse to determine whether the expression expresses a proposition. > Ron Ross said something to the effect that it is very important that a > proposition is one independent of any Universe of Discourse. I think he's > right -- but the current definition is not clear about that. I understand > that the RTF will propose a change to clarify this point. > > 2. Regarding the example given above, what I heard is that whether "John > is in London" may not be known in the model, and is not inferable solely > from the two facts given. So the expression may not be decidable, but it > is still expresses a proposition. > > > Is this a correct summary? > -------------------------------- > Mark H. Linehan > STSM, Model Driven Business Transformation > IBM Research > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov < mailto:edbark@nist.gov> > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > [Ronald G. Ross Contact Information] > http://www.brsolutions.com/email/wordpress-2.png *Blog* || > < http://www.ronross.info/blog/ > > http://www.RonRoss.info/blog/ > http://www.brsolutions.com/email/linkedin.png *LinkedIn* || > < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.brsolutions.com/email/twitter.png < > https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> > *Twitter* || > < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross > http://www.brsolutions.com/email/button-green.png *Homepage* > || < http://www.ronross.info/ > > http://www.RonRoss.info > > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 > Twitter || https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > Twitter || > https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > Date: Fri, 23 Sep 2011 12:07:12 +1000 Subject: Re: meaning of 'proposition' From: Terry Halpin To: Stan Hendryx Cc: "Barkmeyer, Edward J" , "Ronald G. Ross" , sbvr-rtf@omg.org, date-time@omg.org Hi Stan Your rewording is better, but I still don't find your proposal compelling. Your notion of "inclusion" seems counter-intuitive to me, and I don't see what it buys us. One might regard the state of affairs of John being in London now as exemplifying the state of affairs of John being in London at some time during his complete lifetime, but why call this inclusion, and what are you going to use it for? Why not just make the temporal aspects explicit when formalizing things and be done with it? Cheers Terry On Fri, Sep 23, 2011 at 11:08 AM, Stan Hendryx wrote: Terry, To me, too, the tense ordinarily conveys part of the meaning. What I was talking about as the primitive notion that is included is a tenseless concept of, say, John being in London. This primitive notion corresponds to Plantinga.s .haecceity.. Thus, .John is in London. (meaning John is in London now) includes the primitive state of affairs (that might be better stated in English as) .John being in London.. I tried overloading .is. to express the no-tense idea of a haecceity, but that didn.t work; it is too confusing. I.ll switch to using the gerund form to express a haecceity, which seems to be timeless. Plantinga never said how to express a haecceity in English, so I.m making it up. I.ll try again using a gerund: Every proposition that is about John being in London, whenever, corresponds to a state of affairs that includes the state of affairs .John being in London.. Thus, the SOA .John is in London. includes the SOA .John being in London.. So does .John was in London last Thursday.. I formulate .John is in London. as .The time interval of John being in London overlaps the current time.; or ....overlaps last Thursday.; or ....overlaps .. If John is in London often, then the time interval of John being in London overlaps the time interval of each visit, i.e. is the convex hull of the time intervals of all of the visits. We say that .John being in London. .recurs., but of course it doesn.t recur in the literal sense of the word -- no state of affairs literally happens more than once; for a state of affairs to recur means the state of affairs is part of more than one other state of affairs and its time interval overlaps that of each other state of affairs of which it is part. (.SOA2 is part of SOA1. is a synonymous form of .SOA1 includes SOA2., a reverse reading.) I hope this helps clarify what I meant. Best regards, Stan -------------------------------------------------------------------------------- From: Terry Halpin [mailto:terry.halpin@logicblox.com] Sent: Thursday, September 22, 2011 5:15 PM To: Stan Hendryx Cc: Barkmeyer, Edward J; Ronald G. Ross; sbvr-rtf@omg.org; date-time@omg.org Subject: Re: meaning of 'proposition' I don't agree that the sentence 'John is in London' expresses an "atemporal proposition" that corresponds to a "primitive state of affairs". To me, the tense of a sentence conveys part of its meaning. So the present tense sentence 'John is in London' asserts that John is in London at the time the sentence was uttered. Of course, it is possible to combine tenses in a sentence, e.g. 'John is, was, or will be in London' does express a proposition that corresponds to a state of affairs. Cheers Terry On Tue, Sep 20, 2011 at 10:17 AM, Stan Hendryx wrote: The matter of "always true or false (but not both)" is explained by considering that states of affairs denoted by sentences like 'John is in London at least once a month' include the state of affairs denoted by 'John is in London', where the latter is interpreted as being the primitive notion of John being in London, regardless of time. In this context, 'is' does not mean John is in London now, just what it means for John to be in London at all. It is this primitive state of affairs that is said to occur at least once a month, but saying it recurs is just a figure of speech. Formally, such a state of affairs does not literally itself recur; no state of affairs happens more than once. Rather, the primitive state of affairs is included in each state of affairs that is a monthly occurrence. The primitive is what occurs in an occurrence. In Plantinga's theory of actualism, the primitive states of affairs are called "haecceities", which is the essential property of a thing being just what it is. 'Includes' is transitive: if A includes B and B includes C, then A includes C. Thus, the graph of the 'includes' relation in a possible world is a lattice, not a tree, because the primitives can be part of many particular occurrences. The lattice is defined by the set of states of affairs the possible world includes (the set is called the domain of the possible world) and the 'includes' relation. The possible world is the top of the lattice, so it is a sort of maximal state of affairs, being the greatest lower bound of the set under 'includes'. A possible world includes a particular state of affairs at most once, as a member of its domain. The primitive proposition 'John is in London' is true in a possible world just in case the possible world includes at least one state of affairs that includes the primitive state of affairs that John is in London. The primitive is a genuine state of affairs; it occurs for a time interval that is the convex hull of each state of affairs that includes it. Stan > -----Original Message----- > From: Barkmeyer, Edward J [mailto:edward.barkmeyer@nist.gov] > Sent: Monday, September 19, 2011 2:16 PM > To: Terry Halpin; Ronald G. Ross > Cc: Stan Hendryx; sbvr-rtf@omg.org; date-time@omg.org > Subject: RE: meaning of 'proposition' > > Terry wrote: > > > To me, a proposition is what it is that is asserted by a declarative > sentence (or the declarative component of a sentence), > > and is always true or false (but not both). The SBVR "meaning that is > true or false" seems to correspond pretty close to this. > > Well, the 'but not both' is precisely the point of the issue. In 'John is > in London at least once a month', the SBVR approach to modeling this > statement makes 'John is in London' a declarative component of a sentence, > but possibly not what Terry means by 'the declarative component of a > sentence'. If the same fact model includes 'John is not in London on > Fridays', it is clear that the component 'John is in London' cannot be > either true or false, unless it is both. In either case, it fails Terry's > proposed definition of 'proposition' and therefore is not a proposition. > It follows that it does not correspond to any 'state of affairs'. This is > the dilemma that motivated the issue. > > The particular problem here is that Terry's definition is intended to > relate the concept 'proposition' only to sentences in their entirety, i.e., > John is in London at least once a month is a proposition, but no subset of > it is, or at least, no subset is being asked whether it is. That resolves > the issue, but with the same consequence: 'John is in London' is not a > declarative sentence in the fact model, and therefore does not represent a > proposition, and thus cannot correspond to a state of affairs. So, it > doesn't support the desired SBVR semantics either. > > -Ed > > -- > Edward J. Barkmeyer Email: edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 > Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 > ________________________________________ > From: Terry Halpin [terry.halpin@logicblox.com] > Sent: Sunday, September 18, 2011 7:54 PM > To: Ronald G. Ross > Cc: Stan Hendryx; Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org > Subject: Re: meaning of 'proposition' > > Hi Ron > > < categories of proposition (relevant to SBVR) -- (1) existential > propositions, and (2) propositions involving some role? > > The current definition of "proposition:" in SBVR seems inadequate ... > "meaning thatis true or false " ... and this is one source of much > consternation. For example, that definition would seem to admit boolean > variables. I think we really need something better.>> > > I distinguish purely existential propositions (e.g. there is a country > code 'US') from propositions that predicate over individuals (e.g. Ron was > born in the USA). To me, a proposition is what it is that is asserted by a > declarative sentence (or the declarative component of a sentence), and is > always true or false (but not both). The SBVR "meaning that is true or > false" seems to correspond pretty close to this. I don't consider a > variable to be a meaning. > > Cheers > Terry > > > > On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross > > wrote: > At 05:00 PM 9/18/2011, Terry Halpin wrote: > < > Stan wrote: These, too, involve roles. "John Hall exists" is a statement > that represents an instance of the unary fact type 'thing exists'. The > role is 'thing'. The bindable target of the role in the example statement > is "John Hall". The actuality denoted by the statement is John Hall. Note > that SBVR does not include the fact type 'statement denotes state of > affairs', so, using SBVR terms, we would say "the actuality that > corresponds to the proposition expressed by the statement is John Hall." > >> > > I disagree with Stan here. I use "role" in two ways. A "fact role" is a > part played in a fact, and always involves a logical predicate (e.g. The > fact type "Person drives Car" has two roles, one being the role of driving > which is played by a person, and the other being the role of being driven > which is played by a car). A "role type" is a non-rigid type (e.g. > Student), unlike rigid types such as Person or Thing. > > The elementary fact "John Hall is alive" involves an individual and a > unary predicate "is alive", which may be treated as the role of being > alive. The existential fact "John Hall exists" involves an individual and > an existential quantifier. The usual view in logic, with which I concur, > is that existence is not a predicate (and certainly not a first-order > predicate). "Exists" is not the same kind of construct as "is alive", and > it does not involve a role. If we use the individual constant "j" for John > Hall, then "John Hall is alive" may be formalized as "j isAlive" (in > postfix notation), but "John Hall exists" is formalized as "Exists x x = > j". > > Terry, > > Would it be fair to say then that there are two categories of > proposition (relevant to SBVR) -- (1) existential propositions, and (2) > propositions involving some role? > > The current definition of "proposition:" in SBVR seems > inadequate ... "meaning that is true or false " ... and this is one source > of much consternation. For example, that definition would seem to admit > boolean variables. I think we really need something better. > > Thanks, > Ron > > > Cheers > Terry > > On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx > > wrote: > > Ron wrote: > Why would we want to get into 'artificial' languages ... *especially* ones > that are already formally grounded?? > > Machines use different formal languages, so there is a need to translate > between these languages for machines to communicate just as there is a > need to translate between natural languages business people use. > > >>What about existential propositions? e.g.., John Hall exists. > > These, too, involve roles. "John Hall exists" is a statement that > represents an instance of the unary fact type 'thing exists'. The role is > 'thing'. The bindable target of the role in the example statement is "John > Hall". The actuality denoted by the statement is John Hall. Note that SBVR > does not include the fact type 'statement denotes state of affairs', so, > using SBVR terms, we would say "the actuality that corresponds to the > proposition expressed by the statement is John Hall." > > In his landmark paper of 1892, Frege explained that a sign has both a > reference and a sense. In SBVR, I take a statement to be a sign, a state > of affairs to be the reference (denotation) of the sign, and a proposition > to be the sense (meaning) of the sign. I thought it might interest you to > see some of Frege's own words on this (from Max Black's 1960 translation > of the original German): > > "It is natural, now, to think of there being connected with a sign (name, > combination of words, letter), besides that to which the sign refers, > which may be called the reference of the sign, also what I should like to > call the sense of the sign, wherein the mode of presentation is contained. > In our example,...the reference of 'evening star' would be the same as > that of 'morning star', but not the sense. > > "The regular connection between a sign, its sense, and its reference is of > such a kind that to the sign there corresponds a definite sense and to > that in turn a definite reference, while to a given reference (an object) > there does not belong only a single sign. The same sense has different > expressions in different languages or even in the same language...It may > be granted that every grammatically well-formed expression representing a > proper name always has a sense. But this is not to say that to the sense > there also corresponds a reference. The words 'the celestial body most > distant from the Earth' have a sense, but it is very doubtful if they also > have a reference...In grasping a sense, one is not certainly assured of a > reference. > [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) > and says that every grammatically well-formed expression that is not > paradoxical and that does not deny a necessity is certainly assured of a > reference (a state of affairs) in some possible world. If the reference is > included in, is part of, the actual world, the state of affairs is an > actuality. This is not exactly the SBVR notion of an actuality.] > > "If words are used in the ordinary way, what one intends to speak of is > their reference. It can also happen, however, that one wishes to talk > about the words themselves or their sense. This happens, for instance, > when the words of another are quoted..." > --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in > Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), > pp. 25-50 > > Thanks, Ron, for the clarification about the context you mentioned. I > understand from your comment that the idea of a proposition giving truth > conditions is new in the SBVR discussions. If a fact model meets the truth > conditions, the proposition is true for that fact model, for the possible > world modeled by the fact model. What it means for a fact model to meet > the conditions of a proposition is spelled out in the model-theoretic > semantics of SBVR, which, as Ed pointed out, is not well documented. This > is an area for future work on SBVR. The model theory of a language is a > significant part of it's formal grounding, important to interpreting the > language and getting translations correct. Not all 'artificial' languages > have a model theory, e.g. UML does not. > > Stan > > On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" > > wrote: > > >>Good discussion. Some notes inserted like this below. > > Ron > > At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: > Stan Hendryx wrote: > Ed wrote: > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, > > Three SBVR fact types are particularly relevant here: > statement expresses proposition > closed logical formulation formalizes statement > closed logical formulation means proposition > > Only in passing. The proposition (meaning) is the same, whether it is > expressed in English or Structured English or Chinese, or SBVR clause 9 > XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR > is that it doesn't seem to have considered that all of these are > Statements -- expressions representing propositions -- and several of them > are at least as formally grounded as SBVR aspires to be. > > These are related by this rule, which is a composite of several SBVR > necessities: > Each statement expresses exactly one proposition and the statement is > formalized by a closed logical formulation that means just that > proposition. > > A statement in CLIF doesn't need to be formalized by an SBVR closed > logical formulation. It is formalized by the language. Statement means > 'representation of a proposition' -- any representation. > > This should be qualified by saying "each statement that is not paradoxical > expresses...", since paradoxes are nonsensical, not having a meaning, no > logical formulation, e.g. "this sentence is false." Such sentences do not > express propositions, either. > > A closed logical formulation is a representation, a translation, of the > statement it formalizes into the language of predicate logic. > > Well, no. It formalizes it into a language defined in clause 9, 13 and 15 > whose semantics is said to be a dialect of predicate logic. There is no > 'the language of predicate logic'; there are many. And whether the logic > language that Clause 9 introduces has a well-defined formal semantics is > still debatable. None has ever been formally written down. > > A CLIF representation of a proposition is a representation in a standard > language for predicate logic. And you yourself used Quine's mathematical > notation to state propositions. Surely that didn't need to be formalized > in SBVR LRMV to be formalized in the language of predicate logic. > > This is all about thinking that Statement means 'natural language > statement' or perhaps 'Structured English statement', and that idea > pervades SBVR. In Date/Time, we have expressed most of the formal > propositions in SBVR SE, OCL and CLIF, of which the last two are formal > and have formal semantics. In point of fact, we have no tool that > generates the SBVR LRMV closed logical formulations. But hope springs > eternal, at least outside of NIST. > > Ron wrote: > "Truth conditions" is new to me in this context. > > This idea came from G. Frege ca. 1892. His work, which includes the > invention of predicate calculus, is the foundation most subsequent work in > logic and formal semantics. > > >>Next time when I say "this context" I will take the time to be more > clear for you (Stan). "This context" means the SBVR-related e-mails I have > been recently reading regarding the definition of "proposition" in SBVR. > There, better? > > > > Certainly logical formulations should be considered out of bounds for > defining what a proposition is. > > On the contrary, a closed logical formulation that means a proposition > definitively says what the proposition is. SBVR gives a definition of the > concept 'proposition': meaning that is true or false. > > I agree with Ron. The point of all this is that the issue of what a > proposition is does not depend on how it is represented. > > >>Thank you, Ed. That's so obvious it should speak for itself (but of > course it can't). > > > And it certainly doesn't depend on its being represented in an XML form > which no tools generate. 'John is in London' definitively says what the > proposition is, at least to speakers of English. You cannot express it > any more clearly in predicate logic, because that would also require that > you explain that the symbol 'John' is an individual concept (logical > constant) that refers to a specific person, and similarly 'London', and > that the predicate symbol 'lives in' refers to the verb concept that is > designated 'lives in' in English. The formal language is just a grammar > that eliminates certain ambiguities in the interpretation, assuming that > you know what the terminal symbols mean. > > To understand the relationship between fact types and their roles, and > propositions, consider that an expression of an instance of a fact type is > a sentence, an atomic sentence. For example, the sentence "Stan was born > in the US," expresses an instance of the binary fact type 'person was born > in country'. Such a sentence expresses a proposition. The sentence is > formalized by an atomic formulation with role bindings, i.e. it is a > closed logical formulation; I call it a closed atomic formulation (my term, > not SBVR). Each role binding is of a role of the fact type. Each role > binding binds to a bindable target, which is either a bound variable, an > individual (Stan and the US in the example) or an identifying expression. > The closed atomic formulation means the proposition that is expressed by > the sentence. The proposition corresponds to a state of affairs. The state > of affairs is the instance of the fact type. If the state of affairs is an > actuality, the proposition is true. To sum up, an instance of a fact type > is a state of affairs that corresponds to the proposition that is > expressed by the statement of the instance of the fact type. > > I thought SBVR was clearer on this. An instance of a fact type is an > actuality that is understood as involving some things in the roles > specified in the fact type form(s). > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? > > If the statement of the proposition is formal in the SBVR sense (colored), > then the proposition will always involve some role, > > I agree. "Murphy's Law" is an expression that denotes a proposition, and > that exrpression does not involve any thing in any role. But the meaning > that is "Murphy's Law" depends on an intransitive verb concept and > involves the general pronoun 'anything' in the sole role in the unary fact > type 'thing goes wrong'. That is one more reason to stop worrying about > the nature of the expression. > The nature of the proposition as a meaning is to conceptualize a situation > as involving things in roles. I know of no counterexamples. > >>What about existential propositions? e.g.., John Hall exists. ... Some > person exists. > > From a business (not logician's) point of view, these to me to involve > roles too. So I agree ... I can think of no counterexamples. Can anyone > else? > > If not, I believe this is a fundamental characteristic in the meaning of > 'proposition'. It seems right at the heart of what SBVR is about. > > > -Ed > > since every closed logical formulation involves at least one fact type > (verb concept) and it's atomic formulation. We know this because every > statement has a verb and by hypothesis, the verb is formalized. In SBVR, > verbs are formalized as fact type designations, and each fact type has at > least one role. > Stan > > On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > < > mailto:rross@brsolutions.com>> wrote: > > At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: > > > Stan Hendryx wrote: > Keri wrote: > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > Yes. The meaning of a proposition is its truth conditions, what will make > it true or false. These conditions are stated precisely by the closed > logical formulation that formalizes a statement that expresses the > proposition. Truth conditions are independent of any facts. > > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, Stan's position is the basis > for the issue. The nature of the proposition is not that it has a truth > value but rather that it is a set of conditions that determines a truth > value in a given world. The issue as written described that set of > conditions as the 'conceptualization of a situation', but perhaps Stan's > characterization is a bit clearer. Those conditions involve things > playing roles (which, according to SBVR is the nature of a state of > affairs). > The truth value of a proposition, it being true or false, depends on the > facts in each fact model in which it is evaluated. It may be true in some > fact models and false in others. It may be unknown. We know what a > proposition means even if it is false or even if its truth value is > unknown. > I hope this helps clarify what Ron meant. > > Stan is probably right that I misunderstood Ron's point. Ron apparently > meant that the meaning that is a proposition is independent of its truth > value. > > Yes, I meant: By definition, a meaning that is a proposition is > independent of its truth value. > > We fully agree. Does Ron agree with Stan that the nature of the meaning > is the 'truth conditions'? And if we agree, can we write a resolution to > 16526 to say that? > > "Truth conditions" is new to me in this context. Certainly logical > formulations should be considered out of bounds for defining what a > proposition is. Ed, you raised some interesting points in your earlier e- > mail response about roles and fact types with respect to defining > propositions I want to hear discussion about. > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? I say > cautiously because (a) I don't know where that takes things, and (b) I > haven't checked to see if the definitions would become circular. > > Ron > > > -Ed > > > Stan > > > > On Sep 16, 2011, at 10:59 AM, keri > < mailto:keri_ah@mac.com> < < > mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: > > Mark, > > Here are some of the points I captured in my notes. (I don't show that > the bits about "independent of any Universe of Discourse" was part of what > Ron said, but perhaps so and I missed that. From the discussion around > the example proposition statements it did appear that they pertained to > some world/UOD.) > > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > I also captured points such as: > * Being true or false does not depend on (a) if it's known that > it's true or false or (b) that it's interesting that it's true > or false. > * Being true or false is different from ambiguity. > > In regards to the example "John is in London." I have to conclude that the > reason it is a proposition (without knowing whether or not John is > actually in London) is because this statement (of proposition) is /based > on/ the vocabulary's fact type "person is in place" and the further aspect > that a person must be in exactly one place. If this is the meaning that > is understood then it can be seen that any statement about some person > being in some place must be either true or false, without knowing which it > is. (I think that's how it works.) When we exchange examples > propositions we have to have the meaning come from the vocabulary concepts > that the proposition is /based on/. Otherwise, there is no meaning. > > - Keri > > > On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: > > During today's RTF discussion, we discussed this example from Ed's note of > September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My > purpose in writing this email is to summarize what I think I heard. > Please comment if you think I got it wrong. > > > Given a fact model comprising: John is in London during 5-8 September > > 2011. John is in St. Imier during 9-10 September 2011. The time of > > interest in the Universe of Discourse is Sept 5-10, 2011. The > > proposition expressed by 'John is in London' is used in representing > > this fact model, both in SBVR examples, and formally in Date/Time. Is > > it true or false? I believe that in that actual world, the proposition > > is neither true nor false. Therefore, it is a counterexample to what > > Don says. > > 1. SBVR's definition of "proposition" is "meaning that is true or false". > Despite the use of the word "is" in the definition, this definition does > not mean that you have to interpret an expression against a Universe of > Discourse to determine whether the expression expresses a proposition. > Ron Ross said something to the effect that it is very important that a > proposition is one independent of any Universe of Discourse. I think he's > right -- but the current definition is not clear about that. I understand > that the RTF will propose a change to clarify this point. > > 2. Regarding the example given above, what I heard is that whether "John > is in London" may not be known in the model, and is not inferable solely > from the two facts given. So the expression may not be decidable, but it > is still expresses a proposition. > > > Is this a correct summary? > -------------------------------- > Mark H. Linehan > STSM, Model Driven Business Transformation > IBM Research > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov < mailto:edbark@nist.gov> > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > [Ronald G. Ross Contact Information] > http://www.brsolutions.com/email/wordpress-2.png *Blog* || > < http://www.ronross.info/blog/ > > http://www.RonRoss.info/blog/ > http://www.brsolutions.com/email/linkedin.png *LinkedIn* || > < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.brsolutions.com/email/twitter.png < > https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> > *Twitter* || > < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross > http://www.brsolutions.com/email/button-green.png *Homepage* > || < http://www.ronross.info/ > > http://www.RonRoss.info > > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 > Twitter || https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > Twitter || > https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > X-Yahoo-Newman-Id: 825506.79731.bm@omp1025.access.mail.mud.yahoo.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1316757789; bh=1ECK0K8WwhzG8fib5ZXWoadIlb6nr7qEbAjOVfZ1bKw=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:From:To:Cc:References:Subject:Date:Organization:Message-ID:MIME-Version:Content-Type:X-Mailer:X-MimeOLE:Thread-Index:In-Reply-To; b=0TrYsOkusWQPU9EEdCyVv7QkT+JV07LM3ynKmveClTyv3Lw88QE8WtyNylTHKlHp+kHtklyq15ISvkjS9m/M/BV/xTV2zejGYQIKlEViwvjquOPYrky7vgFscqr9800lhFGXc8tL8kDI7dF3dcmxcCfRy56v1JbOUhBlYURC7Yg= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: Nw04LDIVM1nNceKJUCoVjW53tOgIr7Bnbdgb5xKop6Rcm4d gGWevO8Z1mHbAlU_LSmFrnsaBHxVSqDd74ZQmJtlhT2GMf.k_8.Nh9VMJbx0 Yusv84tRTSPDeYiorDoSJ7EwixHhvF0a4meQjV_eIXXZzWSt7p9CAdlvrznB JKW0YMsgm7qsqQTk3iM08J3BMEGgRJtFc1ArsTXPazXegiQd1l9fWsXc4Byi E88RtT1ASeHDKSO9CQtpDF7yOEu6cTH634xA8X.5gwGrZCuoZ15I34Bohyeb MgCKt_6tu8L8XUwr6icQBzL2rwNHKYj5eg06IfqZDuX.9D.lfFgHra41YmGr 0QeLq9NAsCkU8uiOobDMqsiI9GFCcm76YsBftVgvgiy8VxwcY9HBYQAhsYGx 4ZUboHOOl3ZvNYk6kcjEe6PFkNA-- X-Yahoo-SMTP: yZmfYpGswBANaaRY2ZOOLP_anhSkh.0YqFwv79wX8IAI From: "Stan Hendryx" To: "'Terry Halpin'" Cc: "'Barkmeyer, Edward J'" , "'Ronald G. Ross'" , , Subject: RE: meaning of 'proposition' Date: Thu, 22 Sep 2011 23:02:56 -0700 Organization: Hendryx & Associates X-Mailer: Microsoft Office Outlook 11 Thread-Index: Acx5lYN9lmZhMp+gTIO7Gcd0PulsUAAGgj1g Hi Terry, An haecceity is the most primitive notion of a state of affairs, corresponding to the base meaning of a proposition stripped of all modality or qualitative aspects, of which anything can be predicated using different modalities, different times, different places, or different relations. An advantage I see is that, as such, an haecceity provides a point of connection, a nexus, for grouping states of affairs and traversing the inclusion graph of a possible world, which is a lattice of states of affairs defined by .includes. and the set of states of affairs of the possible world, Plantinga.s domain. For example, from the node .John being in London. we can navigate directly to all other states of affairs that include it. We can use those inclusions to reason about all the times John is in London, and about the times John wants to be in London, or wishes he were not in London, or that he must be in London. We can also use it to reason about where he stayed in London, who paid for the trip, what he did . anything that can be predicated of John being in London. An haecceity also provides a link between these properties of John and John himself, and between these properties of London and London itself. Thence from these to all of the other properties and relationships, of any modality, of John and London. I see haecceities as powerful, unifying objects to aid in reasoning about possible worlds with arbitrary modality. This is Davidson.s model of objectification carried to a most general form. Reasoning about a possible world then becomes an exercise in exploring the lattice of the possible world, which is a well-defined algebraic structure. Best regards, Stan -------------------------------------------------------------------------------- From: Terry Halpin [mailto:terry.halpin@logicblox.com] Sent: Thursday, September 22, 2011 7:07 PM To: Stan Hendryx Cc: Barkmeyer, Edward J; Ronald G. Ross; sbvr-rtf@omg.org; date-time@omg.org Subject: Re: meaning of 'proposition' Hi Stan Your rewording is better, but I still don't find your proposal compelling. Your notion of "inclusion" seems counter-intuitive to me, and I don't see what it buys us. One might regard the state of affairs of John being in London now as exemplifying the state of affairs of John being in London at some time during his complete lifetime, but why call this inclusion, and what are you going to use it for? Why not just make the temporal aspects explicit when formalizing things and be done with it? Cheers Terry On Fri, Sep 23, 2011 at 11:08 AM, Stan Hendryx wrote: Terry, To me, too, the tense ordinarily conveys part of the meaning. What I was talking about as the primitive notion that is included is a tenseless concept of, say, John being in London. This primitive notion corresponds to Plantinga.s .haecceity.. Thus, .John is in London. (meaning John is in London now) includes the primitive state of affairs (that might be better stated in English as) .John being in London.. I tried overloading .is. to express the no-tense idea of a haecceity, but that didn.t work; it is too confusing. I.ll switch to using the gerund form to express a haecceity, which seems to be timeless. Plantinga never said how to express a haecceity in English, so I.m making it up. I.ll try again using a gerund: Every proposition that is about John being in London, whenever, corresponds to a state of affairs that includes the state of affairs .John being in London.. Thus, the SOA .John is in London. includes the SOA .John being in London.. So does .John was in London last Thursday.. I formulate .John is in London. as .The time interval of John being in London overlaps the current time.; or ....overlaps last Thursday.; or ....overlaps .. If John is in London often, then the time interval of John being in London overlaps the time interval of each visit, i.e. is the convex hull of the time intervals of all of the visits. We say that .John being in London. .recurs., but of course it doesn.t recur in the literal sense of the word -- no state of affairs literally happens more than once; for a state of affairs to recur means the state of affairs is part of more than one other state of affairs and its time interval overlaps that of each other state of affairs of which it is part. (.SOA2 is part of SOA1. is a synonymous form of .SOA1 includes SOA2., a reverse reading.) I hope this helps clarify what I meant. Best regards, Stan -------------------------------------------------------------------------------- From: Terry Halpin [mailto:terry.halpin@logicblox.com] Sent: Thursday, September 22, 2011 5:15 PM To: Stan Hendryx Cc: Barkmeyer, Edward J; Ronald G. Ross; sbvr-rtf@omg.org; date-time@omg.org Subject: Re: meaning of 'proposition' I don't agree that the sentence 'John is in London' expresses an "atemporal proposition" that corresponds to a "primitive state of affairs". To me, the tense of a sentence conveys part of its meaning. So the present tense sentence 'John is in London' asserts that John is in London at the time the sentence was uttered. Of course, it is possible to combine tenses in a sentence, e.g. 'John is, was, or will be in London' does express a proposition that corresponds to a state of affairs. Cheers Terry On Tue, Sep 20, 2011 at 10:17 AM, Stan Hendryx wrote: The matter of "always true or false (but not both)" is explained by considering that states of affairs denoted by sentences like 'John is in London at least once a month' include the state of affairs denoted by 'John is in London', where the latter is interpreted as being the primitive notion of John being in London, regardless of time. In this context, 'is' does not mean John is in London now, just what it means for John to be in London at all. It is this primitive state of affairs that is said to occur at least once a month, but saying it recurs is just a figure of speech. Formally, such a state of affairs does not literally itself recur; no state of affairs happens more than once. Rather, the primitive state of affairs is included in each state of affairs that is a monthly occurrence. The primitive is what occurs in an occurrence. In Plantinga's theory of actualism, the primitive states of affairs are called "haecceities", which is the essential property of a thing being just what it is. 'Includes' is transitive: if A includes B and B includes C, then A includes C. Thus, the graph of the 'includes' relation in a possible world is a lattice, not a tree, because the primitives can be part of many particular occurrences. The lattice is defined by the set of states of affairs the possible world includes (the set is called the domain of the possible world) and the 'includes' relation. The possible world is the top of the lattice, so it is a sort of maximal state of affairs, being the greatest lower bound of the set under 'includes'. A possible world includes a particular state of affairs at most once, as a member of its domain. The primitive proposition 'John is in London' is true in a possible world just in case the possible world includes at least one state of affairs that includes the primitive state of affairs that John is in London. The primitive is a genuine state of affairs; it occurs for a time interval that is the convex hull of each state of affairs that includes it. Stan > -----Original Message----- > From: Barkmeyer, Edward J [mailto:edward.barkmeyer@nist.gov] > Sent: Monday, September 19, 2011 2:16 PM > To: Terry Halpin; Ronald G. Ross > Cc: Stan Hendryx; sbvr-rtf@omg.org; date-time@omg.org > Subject: RE: meaning of 'proposition' > > Terry wrote: > > > To me, a proposition is what it is that is asserted by a declarative > sentence (or the declarative component of a sentence), > > and is always true or false (but not both). The SBVR "meaning that is > true or false" seems to correspond pretty close to this. > > Well, the 'but not both' is precisely the point of the issue. In 'John is > in London at least once a month', the SBVR approach to modeling this > statement makes 'John is in London' a declarative component of a sentence, > but possibly not what Terry means by 'the declarative component of a > sentence'. If the same fact model includes 'John is not in London on > Fridays', it is clear that the component 'John is in London' cannot be > either true or false, unless it is both. In either case, it fails Terry's > proposed definition of 'proposition' and therefore is not a proposition. > It follows that it does not correspond to any 'state of affairs'. This is > the dilemma that motivated the issue. > > The particular problem here is that Terry's definition is intended to > relate the concept 'proposition' only to sentences in their entirety, i.e., > John is in London at least once a month is a proposition, but no subset of > it is, or at least, no subset is being asked whether it is. That resolves > the issue, but with the same consequence: 'John is in London' is not a > declarative sentence in the fact model, and therefore does not represent a > proposition, and thus cannot correspond to a state of affairs. So, it > doesn't support the desired SBVR semantics either. > > -Ed > > -- > Edward J. Barkmeyer Email: edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Office: +1 301-975-3528 > Gaithersburg, MD 20899-8263 Mobile: +1 240-672-5800 > ________________________________________ > From: Terry Halpin [terry.halpin@logicblox.com] > Sent: Sunday, September 18, 2011 7:54 PM > To: Ronald G. Ross > Cc: Stan Hendryx; Barkmeyer, Edward J; sbvr-rtf@omg.org; date-time@omg.org > Subject: Re: meaning of 'proposition' > > Hi Ron > > < categories of proposition (relevant to SBVR) -- (1) existential > propositions, and (2) propositions involving some role? > > The current definition of "proposition:" in SBVR seems inadequate ... > "meaning thatis true or false " ... and this is one source of much > consternation. For example, that definition would seem to admit boolean > variables. I think we really need something better.>> > > I distinguish purely existential propositions (e.g. there is a country > code 'US') from propositions that predicate over individuals (e.g. Ron was > born in the USA). To me, a proposition is what it is that is asserted by a > declarative sentence (or the declarative component of a sentence), and is > always true or false (but not both). The SBVR "meaning that is true or > false" seems to correspond pretty close to this. I don't consider a > variable to be a meaning. > > Cheers > Terry > > > > On Mon, Sep 19, 2011 at 9:33 AM, Ronald G. Ross > > wrote: > At 05:00 PM 9/18/2011, Terry Halpin wrote: > < > Stan wrote: These, too, involve roles. "John Hall exists" is a statement > that represents an instance of the unary fact type 'thing exists'. The > role is 'thing'. The bindable target of the role in the example statement > is "John Hall". The actuality denoted by the statement is John Hall. Note > that SBVR does not include the fact type 'statement denotes state of > affairs', so, using SBVR terms, we would say "the actuality that > corresponds to the proposition expressed by the statement is John Hall." > >> > > I disagree with Stan here. I use "role" in two ways. A "fact role" is a > part played in a fact, and always involves a logical predicate (e.g. The > fact type "Person drives Car" has two roles, one being the role of driving > which is played by a person, and the other being the role of being driven > which is played by a car). A "role type" is a non-rigid type (e.g. > Student), unlike rigid types such as Person or Thing. > > The elementary fact "John Hall is alive" involves an individual and a > unary predicate "is alive", which may be treated as the role of being > alive. The existential fact "John Hall exists" involves an individual and > an existential quantifier. The usual view in logic, with which I concur, > is that existence is not a predicate (and certainly not a first-order > predicate). "Exists" is not the same kind of construct as "is alive", and > it does not involve a role. If we use the individual constant "j" for John > Hall, then "John Hall is alive" may be formalized as "j isAlive" (in > postfix notation), but "John Hall exists" is formalized as "Exists x x = > j". > > Terry, > > Would it be fair to say then that there are two categories of > proposition (relevant to SBVR) -- (1) existential propositions, and (2) > propositions involving some role? > > The current definition of "proposition:" in SBVR seems > inadequate ... "meaning that is true or false " ... and this is one source > of much consternation. For example, that definition would seem to admit > boolean variables. I think we really need something better. > > Thanks, > Ron > > > Cheers > Terry > > On Sun, Sep 18, 2011 at 5:49 PM, Stan Hendryx > > wrote: > > Ron wrote: > Why would we want to get into 'artificial' languages ... *especially* ones > that are already formally grounded?? > > Machines use different formal languages, so there is a need to translate > between these languages for machines to communicate just as there is a > need to translate between natural languages business people use. > > >>What about existential propositions? e.g.., John Hall exists. > > These, too, involve roles. "John Hall exists" is a statement that > represents an instance of the unary fact type 'thing exists'. The role is > 'thing'. The bindable target of the role in the example statement is "John > Hall". The actuality denoted by the statement is John Hall. Note that SBVR > does not include the fact type 'statement denotes state of affairs', so, > using SBVR terms, we would say "the actuality that corresponds to the > proposition expressed by the statement is John Hall." > > In his landmark paper of 1892, Frege explained that a sign has both a > reference and a sense. In SBVR, I take a statement to be a sign, a state > of affairs to be the reference (denotation) of the sign, and a proposition > to be the sense (meaning) of the sign. I thought it might interest you to > see some of Frege's own words on this (from Max Black's 1960 translation > of the original German): > > "It is natural, now, to think of there being connected with a sign (name, > combination of words, letter), besides that to which the sign refers, > which may be called the reference of the sign, also what I should like to > call the sense of the sign, wherein the mode of presentation is contained. > In our example,...the reference of 'evening star' would be the same as > that of 'morning star', but not the sense. > > "The regular connection between a sign, its sense, and its reference is of > such a kind that to the sign there corresponds a definite sense and to > that in turn a definite reference, while to a given reference (an object) > there does not belong only a single sign. The same sense has different > expressions in different languages or even in the same language...It may > be granted that every grammatically well-formed expression representing a > proper name always has a sense. But this is not to say that to the sense > there also corresponds a reference. The words 'the celestial body most > distant from the Earth' have a sense, but it is very doubtful if they also > have a reference...In grasping a sense, one is not certainly assured of a > reference. > [The modern view adopts possible world semantics (Saul Kripke, ca. 1960) > and says that every grammatically well-formed expression that is not > paradoxical and that does not deny a necessity is certainly assured of a > reference (a state of affairs) in some possible world. If the reference is > included in, is part of, the actual world, the state of affairs is an > actuality. This is not exactly the SBVR notion of an actuality.] > > "If words are used in the ordinary way, what one intends to speak of is > their reference. It can also happen, however, that one wishes to talk > about the words themselves or their sense. This happens, for instance, > when the words of another are quoted..." > --G. Frege, "Ueber Sinn und Bedeutung (On Sense And Reference)," in > Zeitschrift fuer Philosophie und philosophische Kritik, vol. 100 (1892), > pp. 25-50 > > Thanks, Ron, for the clarification about the context you mentioned. I > understand from your comment that the idea of a proposition giving truth > conditions is new in the SBVR discussions. If a fact model meets the truth > conditions, the proposition is true for that fact model, for the possible > world modeled by the fact model. What it means for a fact model to meet > the conditions of a proposition is spelled out in the model-theoretic > semantics of SBVR, which, as Ed pointed out, is not well documented. This > is an area for future work on SBVR. The model theory of a language is a > significant part of it's formal grounding, important to interpreting the > language and getting translations correct. Not all 'artificial' languages > have a model theory, e.g. UML does not. > > Stan > > On Sep 17, 2011, at 3:45 PM, "Ronald G. Ross" > > wrote: > > >>Good discussion. Some notes inserted like this below. > > Ron > > At 08:20 PM 9/16/2011, Edward Barkmeyer wrote: > Stan Hendryx wrote: > Ed wrote: > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, > > Three SBVR fact types are particularly relevant here: > statement expresses proposition > closed logical formulation formalizes statement > closed logical formulation means proposition > > Only in passing. The proposition (meaning) is the same, whether it is > expressed in English or Structured English or Chinese, or SBVR clause 9 > XML, or CLIF or OCL (as in Date/Time). One of the many annoyances of SBVR > is that it doesn't seem to have considered that all of these are > Statements -- expressions representing propositions -- and several of them > are at least as formally grounded as SBVR aspires to be. > > These are related by this rule, which is a composite of several SBVR > necessities: > Each statement expresses exactly one proposition and the statement is > formalized by a closed logical formulation that means just that > proposition. > > A statement in CLIF doesn't need to be formalized by an SBVR closed > logical formulation. It is formalized by the language. Statement means > 'representation of a proposition' -- any representation. > > This should be qualified by saying "each statement that is not paradoxical > expresses...", since paradoxes are nonsensical, not having a meaning, no > logical formulation, e.g. "this sentence is false." Such sentences do not > express propositions, either. > > A closed logical formulation is a representation, a translation, of the > statement it formalizes into the language of predicate logic. > > Well, no. It formalizes it into a language defined in clause 9, 13 and 15 > whose semantics is said to be a dialect of predicate logic. There is no > 'the language of predicate logic'; there are many. And whether the logic > language that Clause 9 introduces has a well-defined formal semantics is > still debatable. None has ever been formally written down. > > A CLIF representation of a proposition is a representation in a standard > language for predicate logic. And you yourself used Quine's mathematical > notation to state propositions. Surely that didn't need to be formalized > in SBVR LRMV to be formalized in the language of predicate logic. > > This is all about thinking that Statement means 'natural language > statement' or perhaps 'Structured English statement', and that idea > pervades SBVR. In Date/Time, we have expressed most of the formal > propositions in SBVR SE, OCL and CLIF, of which the last two are formal > and have formal semantics. In point of fact, we have no tool that > generates the SBVR LRMV closed logical formulations. But hope springs > eternal, at least outside of NIST. > > Ron wrote: > "Truth conditions" is new to me in this context. > > This idea came from G. Frege ca. 1892. His work, which includes the > invention of predicate calculus, is the foundation most subsequent work in > logic and formal semantics. > > >>Next time when I say "this context" I will take the time to be more > clear for you (Stan). "This context" means the SBVR-related e-mails I have > been recently reading regarding the definition of "proposition" in SBVR. > There, better? > > > > Certainly logical formulations should be considered out of bounds for > defining what a proposition is. > > On the contrary, a closed logical formulation that means a proposition > definitively says what the proposition is. SBVR gives a definition of the > concept 'proposition': meaning that is true or false. > > I agree with Ron. The point of all this is that the issue of what a > proposition is does not depend on how it is represented. > > >>Thank you, Ed. That's so obvious it should speak for itself (but of > course it can't). > > > And it certainly doesn't depend on its being represented in an XML form > which no tools generate. 'John is in London' definitively says what the > proposition is, at least to speakers of English. You cannot express it > any more clearly in predicate logic, because that would also require that > you explain that the symbol 'John' is an individual concept (logical > constant) that refers to a specific person, and similarly 'London', and > that the predicate symbol 'lives in' refers to the verb concept that is > designated 'lives in' in English. The formal language is just a grammar > that eliminates certain ambiguities in the interpretation, assuming that > you know what the terminal symbols mean. > > To understand the relationship between fact types and their roles, and > propositions, consider that an expression of an instance of a fact type is > a sentence, an atomic sentence. For example, the sentence "Stan was born > in the US," expresses an instance of the binary fact type 'person was born > in country'. Such a sentence expresses a proposition. The sentence is > formalized by an atomic formulation with role bindings, i.e. it is a > closed logical formulation; I call it a closed atomic formulation (my term, > not SBVR). Each role binding is of a role of the fact type. Each role > binding binds to a bindable target, which is either a bound variable, an > individual (Stan and the US in the example) or an identifying expression. > The closed atomic formulation means the proposition that is expressed by > the sentence. The proposition corresponds to a state of affairs. The state > of affairs is the instance of the fact type. If the state of affairs is an > actuality, the proposition is true. To sum up, an instance of a fact type > is a state of affairs that corresponds to the proposition that is > expressed by the statement of the instance of the fact type. > > I thought SBVR was clearer on this. An instance of a fact type is an > actuality that is understood as involving some things in the roles > specified in the fact type form(s). > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? > > If the statement of the proposition is formal in the SBVR sense (colored), > then the proposition will always involve some role, > > I agree. "Murphy's Law" is an expression that denotes a proposition, and > that exrpression does not involve any thing in any role. But the meaning > that is "Murphy's Law" depends on an intransitive verb concept and > involves the general pronoun 'anything' in the sole role in the unary fact > type 'thing goes wrong'. That is one more reason to stop worrying about > the nature of the expression. > The nature of the proposition as a meaning is to conceptualize a situation > as involving things in roles. I know of no counterexamples. > >>What about existential propositions? e.g.., John Hall exists. ... Some > person exists. > > From a business (not logician's) point of view, these to me to involve > roles too. So I agree ... I can think of no counterexamples. Can anyone > else? > > If not, I believe this is a fundamental characteristic in the meaning of > 'proposition'. It seems right at the heart of what SBVR is about. > > > -Ed > > since every closed logical formulation involves at least one fact type > (verb concept) and it's atomic formulation. We know this because every > statement has a verb and by hypothesis, the verb is formalized. In SBVR, > verbs are formalized as fact type designations, and each fact type has at > least one role. > Stan > > On Sep 16, 2011, at 3:28 PM, "Ronald G. Ross" > < > mailto:rross@brsolutions.com>> wrote: > > At 03:46 PM 9/16/2011, Ed Barkmeyer wrote: > > > Stan Hendryx wrote: > Keri wrote: > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > Yes. The meaning of a proposition is its truth conditions, what will make > it true or false. These conditions are stated precisely by the closed > logical formulation that formalizes a statement that expresses the > proposition. Truth conditions are independent of any facts. > > Ignoring the idea that a proposition has to be stated as an SBVR closed > logical formulation in order to be expressed, Stan's position is the basis > for the issue. The nature of the proposition is not that it has a truth > value but rather that it is a set of conditions that determines a truth > value in a given world. The issue as written described that set of > conditions as the 'conceptualization of a situation', but perhaps Stan's > characterization is a bit clearer. Those conditions involve things > playing roles (which, according to SBVR is the nature of a state of > affairs). > The truth value of a proposition, it being true or false, depends on the > facts in each fact model in which it is evaluated. It may be true in some > fact models and false in others. It may be unknown. We know what a > proposition means even if it is false or even if its truth value is > unknown. > I hope this helps clarify what Ron meant. > > Stan is probably right that I misunderstood Ron's point. Ron apparently > meant that the meaning that is a proposition is independent of its truth > value. > > Yes, I meant: By definition, a meaning that is a proposition is > independent of its truth value. > > We fully agree. Does Ron agree with Stan that the nature of the meaning > is the 'truth conditions'? And if we agree, can we write a resolution to > 16526 to say that? > > "Truth conditions" is new to me in this context. Certainly logical > formulations should be considered out of bounds for defining what a > proposition is. Ed, you raised some interesting points in your earlier e- > mail response about roles and fact types with respect to defining > propositions I want to hear discussion about. > > I ask this very. very cautiously: Can anybody think of an example > proposition in the SBVR sense that wouldn't involve some role? I say > cautiously because (a) I don't know where that takes things, and (b) I > haven't checked to see if the definitions would become circular. > > Ron > > > -Ed > > > Stan > > > > On Sep 16, 2011, at 10:59 AM, keri > < mailto:keri_ah@mac.com> < < > mailto:keri_ah@mac.com > mailto:keri_ah@mac.com>> wrote: > > Mark, > > Here are some of the points I captured in my notes. (I don't show that > the bits about "independent of any Universe of Discourse" was part of what > Ron said, but perhaps so and I missed that. From the discussion around > the example proposition statements it did appear that they pertained to > some world/UOD.) > > What I got for Ron's statement was: It's an important feature of SBVR > that a proposition is what it is, *independent of knowing any facts*. > > I also captured points such as: > * Being true or false does not depend on (a) if it's known that > it's true or false or (b) that it's interesting that it's true > or false. > * Being true or false is different from ambiguity. > > In regards to the example "John is in London." I have to conclude that the > reason it is a proposition (without knowing whether or not John is > actually in London) is because this statement (of proposition) is /based > on/ the vocabulary's fact type "person is in place" and the further aspect > that a person must be in exactly one place. If this is the meaning that > is understood then it can be seen that any statement about some person > being in some place must be either true or false, without knowing which it > is. (I think that's how it works.) When we exchange examples > propositions we have to have the meaning come from the vocabulary concepts > that the proposition is /based on/. Otherwise, there is no meaning. > > - Keri > > > On Sep 16, 2011, at 9:46 AM, Mark H Linehan wrote: > > During today's RTF discussion, we discussed this example from Ed's note of > September 9 at 1:10pm, titled "*Re: issue 16526 -- SBVR RTF issue*". My > purpose in writing this email is to summarize what I think I heard. > Please comment if you think I got it wrong. > > > Given a fact model comprising: John is in London during 5-8 September > > 2011. John is in St. Imier during 9-10 September 2011. The time of > > interest in the Universe of Discourse is Sept 5-10, 2011. The > > proposition expressed by 'John is in London' is used in representing > > this fact model, both in SBVR examples, and formally in Date/Time. Is > > it true or false? I believe that in that actual world, the proposition > > is neither true nor false. Therefore, it is a counterexample to what > > Don says. > > 1. SBVR's definition of "proposition" is "meaning that is true or false". > Despite the use of the word "is" in the definition, this definition does > not mean that you have to interpret an expression against a Universe of > Discourse to determine whether the expression expresses a proposition. > Ron Ross said something to the effect that it is very important that a > proposition is one independent of any Universe of Discourse. I think he's > right -- but the current definition is not clear about that. I understand > that the RTF will propose a change to clarify this point. > > 2. Regarding the example given above, what I heard is that whether "John > is in London" may not be known in the model, and is not inferable solely > from the two facts given. So the expression may not be decidable, but it > is still expresses a proposition. > > > Is this a correct summary? > -------------------------------- > Mark H. Linehan > STSM, Model Driven Business Transformation > IBM Research > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov < mailto:edbark@nist.gov> > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > [Ronald G. Ross Contact Information] > http://www.brsolutions.com/email/wordpress-2.png *Blog* || > < http://www.ronross.info/blog/ > > http://www.RonRoss.info/blog/ > http://www.brsolutions.com/email/linkedin.png *LinkedIn* || > < http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > http://www.brsolutions.com/email/twitter.png < > https://twitter.com/Ronald_G_Ross> < https://twitter.com/Ronald_G_Ross> > *Twitter* || > < https://twitter.com/Ronald_G_Ross > https://twitter.com/Ronald_G_Ross > http://www.brsolutions.com/email/button-green.png *Homepage* > || < http://www.ronross.info/ > > http://www.RonRoss.info > > > -- > Edward J. Barkmeyer Email: > edbark@nist.gov > National Institute of Standards & Technology > Manufacturing Systems Integration Division > 100 Bureau Drive, Stop 8263 Tel: +1 301-975- > 3528 > Gaithersburg, MD 20899-8263 Cel: +1 240-672- > 5800 > > "The opinions expressed above do not reflect consensus of NIST, and have > not been reviewed by any Government authority." > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || http://www.linkedin.com/pub/ronald-ross/1/3b/346 > Twitter || https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > > > Blog || http://www.RonRoss.info/blog/ > LinkedIn || > http://www.linkedin.com/pub/ronald-ross/1/3b/346 > > Twitter || > https://twitter.com/Ronald_G_Ross > Homepage || http://www.RonRoss.info > > From: "Donald Chapin" To: "sbvr-rtf " Subject: FW: issue 16526 -- SBVR RTF issue Date: Tue, 18 Oct 2011 15:20:15 +0100 X-Mailer: Microsoft Outlook 14.0 Thread-Index: AQKwu/F+6yIIN2qth/DDIQ+f11yanpO5yG8w X-Mirapoint-IP-Reputation: reputation=Good-1, source=Queried, refid=tid=0001.0A0B0302.4E9D8B21.0090, actions=TAG X-Junkmail-Premium-Raw: score=9/50, refid=2.7.2:2011.10.18.135114:17:9.975, ip=81.149.51.65, rules=__TO_MALFORMED_2, __BOUNCE_NDR_SUBJ_EXEMPT, __SUBJ_ALPHA_END, __HAS_MSGID, __SANE_MSGID, __MIME_VERSION, __CT, __CTYPE_HAS_BOUNDARY, __CTYPE_MULTIPART, __CTYPE_MULTIPART_MIXED, __HAS_X_MAILER, __OUTLOOK_MUA_1, __USER_AGENT_MS_GENERIC, DOC_ATTACHED, __ANY_URI, LINK_TO_IMAGE, __FRAUD_CONTACT_NUM, __STOCK_PHRASE_24, __CP_URI_IN_BODY, __C230066_P5, __HTML_MSWORD, __HTML_FONT_BLUE, __HAS_HTML, BODY_SIZE_10000_PLUS, BODYTEXTH_SIZE_10000_LESS, __MIME_HTML, __IMGSPAM_BODY, __TAG_EXISTS_HTML, __STYLE_RATWARE_2, RDNS_GENERIC_POOLED, HTML_50_70, RDNS_SUSP_GENERIC, __OUTLOOK_MUA, RDNS_SUSP, FORGED_MUA_OUTLOOK, IMGSPAM_BODY X-Junkmail-Status: score=10/50, host=c2beaomr06.btconnect.com X-Junkmail-Signature-Raw: score=unknown, refid=str=0001.0A0B0202.4E9D8BBB.000C,ss=1,vtr=str,vl=0,fgs=0, ip=0.0.0.0, so=2010-07-22 22:03:31, dmn=2009-09-10 00:05:08, mode=multiengine X-Junkmail-IWF: false A slightly updated draft resolution for Issue 16526 is attached for discussion this week. From: Juergen Boldt [mailto:juergen@omg.org] Sent: 01 September 2011 15:11 To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org Draft Resolution for Issue 16526 (2011-10-18).doc To: sbvr-rtf@omg.org Subject: Re: FW: issue 16526 -- SBVR RTF issue X-KeepSent: 2D67CD73:CC5C5B5E-8525792E:000D0909; type=4; name=$KeepSent X-Mailer: Lotus Notes Release 8.5.1FP5 SHF29 November 12, 2010 From: Mark H Linehan Date: Tue, 18 Oct 2011 22:26:46 -0400 X-MIMETrack: Serialize by Router on D01MC604/01/M/IBM(Release 8.5.2FP1 ZX852FP1HF6|May 2, 2011) at 10/18/2011 22:26:46, Serialize complete at 10/18/2011 22:26:46 x-cbid: 11101902-1976-0000-0000-00000030F8F9 I think the definition and first Note of "proposition" should reflect the point made in the last note for "proposition is true" -- that actuality and truth are with respect to a possible world. Regarding the proposed reference scheme: this means that two different statements always identify different propositions even if they mean the same thing. Example: (1) London is the capital of the United Kingdom; (2) the capital of the United Kingdom is London. In the existing definition, given that these two statements would be formulated the same way, they would be the same proposition. With the proposed reference scheme, they are different propositions. -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research From: "Donald Chapin" To: "sbvr-rtf " Date: 10/18/2011 10:37 AM Subject: FW: issue 16526 -- SBVR RTF issue -------------------------------------------------------------------------------- A slightly updated draft resolution for Issue 16526 is attached for discussion this week. From: Juergen Boldt [mailto:juergen@omg.org] Sent: 01 September 2011 15:11 To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org [attachment "Draft Resolution for Issue 16526 (2011-10-18).doc" deleted by Mark H Linehan/Watson/IBM] From: "Donald Chapin" To: "'Mark H Linehan'" , Subject: RE: FW: issue 16526 -- SBVR RTF issue Date: Wed, 19 Oct 2011 12:04:37 +0100 X-Mailer: Microsoft Outlook 14.0 Thread-Index: AQKwu/F+6yIIN2qth/DDIQ+f11yangJ/ezQ0AX9FqSuTmy6IMA== X-Mirapoint-IP-Reputation: reputation=Good-1, source=Queried, refid=tid=0001.0A0B0303.4E9EAEC8.003E, actions=tag X-Junkmail-Premium-Raw: score=7/50, refid=2.7.2:2011.10.19.93017:17:7.944, ip=81.149.51.65, rules=__TO_MALFORMED_2, __BOUNCE_CHALLENGE_SUBJ, __BOUNCE_NDR_SUBJ_EXEMPT, __SUBJ_ALPHA_END, __HAS_MSGID, __SANE_MSGID, __MIME_VERSION, __CT, __CTYPE_HAS_BOUNDARY, __CTYPE_MULTIPART, __CTYPE_MULTIPART_MIXED, __HAS_X_MAILER, __OUTLOOK_MUA_1, __USER_AGENT_MS_GENERIC, DOC_ATTACHED, __ANY_URI, __FRAUD_CONTACT_NUM, __STOCK_PHRASE_24, __CP_URI_IN_BODY, __C230066_P5, __HTML_MSWORD, __HTML_FONT_BLUE, __HAS_HTML, BODY_SIZE_10000_PLUS, __MIME_HTML, __TAG_EXISTS_HTML, __STYLE_RATWARE_2, RDNS_GENERIC_POOLED, HTML_70_90, RDNS_SUSP_GENERIC, __OUTLOOK_MUA, RDNS_SUSP, FORGED_MUA_OUTLOOK X-Junkmail-Status: score=10/50, host=c2bthomr09.btconnect.com X-Junkmail-Signature-Raw: score=unknown, refid=str=0001.0A0B0209.4E9EAECB.00E5,ss=1,vtr=str,vl=0,fgs=0, ip=0.0.0.0, so=2010-07-22 22:03:31, dmn=2009-09-10 00:05:08, mode=multiengine X-Junkmail-IWF: false Attached is a slightly updated draft resolution to deal with the first of Mark.s points below and with feedback that we omitted updating the fifth necessity in Subclause 8.6.2. Mark, to the best of my knowledge with respect to your second paragraph, there is nothing in the SBVR specification that says that two different identifiers in a reference scheme must identify two different things. Donald From: Mark H Linehan [mailto:mlinehan@us.ibm.com] Sent: 19 October 2011 03:27 To: sbvr-rtf@omg.org Subject: Re: FW: issue 16526 -- SBVR RTF issue I think the definition and first Note of "proposition" should reflect the point made in the last note for "proposition is true" -- that actuality and truth are with respect to a possible world. Regarding the proposed reference scheme: this means that two different statements always identify different propositions even if they mean the same thing. Example: (1) London is the capital of the United Kingdom; (2) the capital of the United Kingdom is London. In the existing definition, given that these two statements would be formulated the same way, they would be the same proposition. With the proposed reference scheme, they are different propositions. -------------------------------- Mark H. Linehan STSM, Model Driven Business Transformation IBM Research From: "Donald Chapin" To: "sbvr-rtf " Date: 10/18/2011 10:37 AM Subject: FW: issue 16526 -- SBVR RTF issue -------------------------------------------------------------------------------- A slightly updated draft resolution for Issue 16526 is attached for discussion this week. From: Juergen Boldt [mailto:juergen@omg.org] Sent: 01 September 2011 15:11 To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org [attachment "Draft Resolution for Issue 16526 (2011-10-18).doc" deleted by Mark H Linehan/Watson/IBM] Draft Resolution for Issue 16526 (2011-10-19).doc Draft Resolution for Issue 16526 (2011-10-19) CHANGES ACCEPTED.doc X-SpamScore: -33 X-BigFish: PS-33(z21cILz709fK936eKc85fh1135M12d5M9a6kzz1202hzz8275bh8275dhz31h2a8h668h839h34h) X-Forefront-Antispam-Report: CIP:207.46.4.139;KIP:(null);UIP:(null);IPV:SKI;H:SN2PRD0302HT010.namprd03.prod.outlook.com;R:internal;EFV:INT From: Don Baisley To: "sbvr-rtf@omg.org" Subject: RE: FW: issue 16526 -- SBVR RTF issue Thread-Topic: FW: issue 16526 -- SBVR RTF issue Thread-Index: AQHMaLFXF0y3f6q7s0yD7Sl4Xh8oHJWCcLmAgADK/ACAAJCwgIACsAEA Date: Fri, 21 Oct 2011 04:10:47 +0000 Accept-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [76.104.188.194] X-OrganizationHeadersPreserved: SN2PRD0302HT010.namprd03.prod.outlook.com X-FOPE-CONNECTOR: Id%0$Dn%*$RO%0$TLS%0$FQDN%$TlsDn% X-FOPE-CONNECTOR: Id%59$Dn%OMG.ORG$RO%2$TLS%6$FQDN%131.107.125.5$TlsDn% X-OriginatorOrg: microsoft.com X-CrossPremisesHeadersPromoted: TK5EX14HUBC104.redmond.corp.microsoft.com X-CrossPremisesHeadersFiltered: TK5EX14HUBC104.redmond.corp.microsoft.com The attached resolution to issue 16526 is based on discussion at last Wednesday.s RTF call and is ready for ballot. Regards, Don -------------------------------------------------------------------------------- From: Juergen Boldt [mailto:juergen@omg.org] Sent: 01 September 2011 15:11 To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org [attachment "Draft Resolution for Issue 16526 (2011-10-18).doc" deleted by Mark H Linehan/Watson/IBM] Issue 16526.doc X-SpamScore: -37 X-BigFish: PS-37(z21cILz709fK9371K936eKc85fh1454I1135M12d5M9a6kzz1202hzz8275bh8275dhz31h2a8h668h839h34h) X-Forefront-Antispam-Report: CIP:207.46.4.139;KIP:(null);UIP:(null);IPV:SKI;H:SN2PRD0302HT003.namprd03.prod.outlook.com;R:internal;EFV:INT From: Don Baisley To: "sbvr-rtf@omg.org" Subject: RE: FW: issue 16526 -- SBVR RTF issue Thread-Topic: FW: issue 16526 -- SBVR RTF issue Thread-Index: AQHMaLFXF0y3f6q7s0yD7Sl4Xh8oHJWCcLmAgADK/ACAAJCwgIACsAEAgArzy/A= Date: Fri, 28 Oct 2011 03:43:31 +0000 Accept-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [76.104.188.194] X-OrganizationHeadersPreserved: SN2PRD0302HT003.namprd03.prod.outlook.com X-FOPE-CONNECTOR: Id%0$Dn%*$RO%0$TLS%0$FQDN%$TlsDn% X-FOPE-CONNECTOR: Id%59$Dn%OMG.ORG$RO%2$TLS%6$FQDN%131.107.125.5$TlsDn% X-OriginatorOrg: microsoft.com X-CrossPremisesHeadersPromoted: TK5EX14HUBC105.redmond.corp.microsoft.com X-CrossPremisesHeadersFiltered: TK5EX14HUBC105.redmond.corp.microsoft.com I was uncomfortable with adding the text about .possible worlds. into the definition of .proposition.. It seems to me to be out of band. Looking at various online dictionaries, I find no definition of .proposition. that involves the concept of .possible worlds.. I asked Terry Halpin. Here is his reply about the .possible world. stuff in the definition: I don't think it helps to include "possible world" in defining "proposition". For example, when I utter the false proposition 1 +1 = 3, it seems strange to bring in a possible world in which it is false to clarify what I am doing. Feel free to pass on my comments. Cheers Terry I would like for the RTF to consider dropping the .possible world. parts that had been proposed for addition into the entry for .proposition.. I attached a revised document with change tracking on. Note that possibility with respect to propositions is addressed separately in another entry in the same SBVR subclause. Also, I fixed a minor problem: the .Resolution. section had not been updated to match the change we made to the reference scheme. Best regards, Don From: Don Baisley [mailto:Don.Baisley@microsoft.com] Sent: Thursday, October 20, 2011 9:11 PM To: sbvr-rtf@omg.org Subject: RE: FW: issue 16526 -- SBVR RTF issue The attached resolution to issue 16526 is based on discussion at last Wednesday.s RTF call and is ready for ballot. Regards, Don -------------------------------------------------------------------------------- From: Juergen Boldt [mailto:juergen@omg.org] Sent: 01 September 2011 15:11 To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org [attachment "Draft Resolution for Issue 16526 (2011-10-18).doc" deleted by Mark H Linehan/Watson/IBM] Issue 16526 (2).doc X-SpamScore: -33 X-BigFish: PS-33(z21cILz709fK936eKc85fh1135M12d5M9a6kzz1202hzz8275bh8275dhz31h2a8h668h839h34h) X-Forefront-Antispam-Report: CIP:207.46.4.139;KIP:(null);UIP:(null);IPV:SKI;H:SN2PRD0302HT004.namprd03.prod.outlook.com;R:internal;EFV:INT From: Don Baisley To: "sbvr-rtf@omg.org" Subject: RE: FW: issue 16526 -- SBVR RTF issue Thread-Topic: FW: issue 16526 -- SBVR RTF issue Thread-Index: AQHMaLFXF0y3f6q7s0yD7Sl4Xh8oHJWCcLmAgADK/ACAAJCwgIACsAEAgArzy/CABCpVQA== Date: Sun, 30 Oct 2011 19:00:46 +0000 Accept-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [76.104.188.194] X-OrganizationHeadersPreserved: SN2PRD0302HT004.namprd03.prod.outlook.com X-FOPE-CONNECTOR: Id%0$Dn%*$RO%0$TLS%0$FQDN%$TlsDn% X-FOPE-CONNECTOR: Id%59$Dn%OMG.ORG$RO%2$TLS%6$FQDN%131.107.125.5$TlsDn% X-OriginatorOrg: microsoft.com X-CrossPremisesHeadersPromoted: TK5EX14MLTC101.redmond.corp.microsoft.com X-CrossPremisesHeadersFiltered: TK5EX14MLTC101.redmond.corp.microsoft.com The attached resolution to issue 16526 includes the changes made in last Friday.s RTF call and is ready for ballot. Regards, Don -------------------------------------------------------------------------------- From: Juergen Boldt [mailto:juergen@omg.org] Sent: 01 September 2011 15:11 To: issues@omg.org; sbvr-rtf@omg.org Subject: issue 16526 -- SBVR RTF issue Date: Wed, 31 Aug 2011 18:53:59 -0400 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: issues@omg.org CC: SBVR RTF Subject: SBVR issue: Definition of proposition X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: p7VMs4Sd022365 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1315436045.00761@Uvt4MxKhHebD5J8M3lIs+w X-Spam-Status: No X-NIST-MailScanner: Found to be clean X-NIST-MailScanner-From: edbark@nist.gov Title: Definition of proposition Specification: SBVR Version: 1.0 Clause: 8.1.2 Source: Ed Barkmeyer, NIST, edbark@nist.gov (for the Date/Time submission team) Summary: SBVR clause 8.1.2 defines 'proposition' as 'meaning that is true or false'. The Date/Time specification, and some SBVR examples, show that some propositions are used for their "content" -- the situation that the proposition describes -- without regard to their truth value. For example, "Each rental car must be inspected before it is available for rental" uses the proposition 'rental car r is inspected' (for each referent of r) to refer to situation in which the car is inspected, and the proposition 'rental car r is available for rental' to refer to the situation in which the car can be rented. The rule relates these situations without requiring any true/false evaluation of either of them. Further, the situation in which a given rental car is available is only sometimes an actuality; the proposition 'r is available for rental' can be sometimes true and sometimes false in the actual world. Thus, being true or false is not the most important characteristic of a proposition, and may not be well-defined. Recommendation: 'proposition' should be defined as: conceptualization of an event, activity, situation or circumstance. Such a definition would be consistent with the idea that it 'corresponds to' a 'state of affairs'. It is also consistent with the idea that true and false are defined in terms of correspondence to an actuality. Those properties would be dependent on the situation that is identified in the proposed definition. This change of definition does not change the intent of the term 'proposition' in any way. It just avoids having the concept depend on having a truth value in usages that don't care. (It may be that the proposed definition needs some additional characteristic to distinguish it from a noun concept that corresponds to events, like 'heart attack'. For example, the proposition must be based on one or more fact types and involve things in fact type roles.) Juergen Boldt Director, Member Services 140 Kendrick Street, Building A Suite 300 Needham, MA 02494 USA Tel: 781 444 0404 x 132 fax: 781 444 0320 www.omg.org [attachment "Draft Resolution for Issue 16526 (2011-10-18).doc" deleted by Mark H Linehan/Watson/IBM] Issue 165261.doc