Hesperus is Bosphorus

A group blog by philosophers in and from Turkey

Wenglish: A Language with No Sentences

with 39 comments

It has been more than a year since I have been working on the idea that truth is in fact a form of reference. I started writing a text last year around this time, intending it to be a journal article, but then it got so long that I am now thinking of turning it into a book. The idea first came up when I was working on one of the chapters of my book on curiosity which just recently came out. I hold that being curious requires one to attempt to refer to something unknown to him/her. This allowed me to deal with the wh-questions easily but I had a serious problem with direct questions, or better yes/no questions. Initially I used Frege’s theory to tackle with it but it was too artificial. I liked the Fregean idea that sentences are in fact referring expressions, but I could not convince myself that true sentences refer to the True (whatever that may be)– and even worse is that false sentences refer to the False. So I started searching for an alternative theory which is what led to this work. I then revised and made substantial additions to my curiosity book, but it was at best scratching the surface. In the past year or so I gave four separate talks on it, originally with the title *TRUTH IS REFERENCE*, in Virginia, Milan, St Andrews, and Bogazici.  The part of the talk that attracted the most amount of attention is where I develop a hypothetical language that I call *Wenglish*. This is a language which is just like English except that it does not have declarative sentences. Well that’s what I say, and though most of my listeners seemed to agree with me someone in the audience in one of my talks objected to it (I think it was in Virginia and it might have been Trenton Merrics, but I have to check this).  Wenglish also does not have a separate truth predicate, but of course it has the notion of *reference*. *Reference* is not a predicate though, because Wenglish does not have any predicates either. Rather it has descriptional functions that do same job.  Anyway I argue that whatever that we can say in English we can say in Wenglish. If this is correct it shows all three things that I wish to show: truth is a form of reference; to say that a sentence is true is to say that it refers; and to say that a sentence is false is to say that it fails to refer. Anyway here is a short passage in Wenglish for you to figure out how it works:

Now the existence of our fine blog.  Lucas’ insisting for me writing on our blog. My telling him my being too busy. My liking the blog. My being surprised to seeing so many entries in a short time. Anyway, the interestingness of the discussions in the blog; there being witty and even funny comments as well. Your understanding how Wenglish working? Wenglish’s having no sentences. Wenglish’s rather having *wentences*. Some people’s objecting. My not caring too much. Just joking. Wenglish’s having no truth predicate. Here is how saying *is true*: “The world’s being round” referring.  Tarskian scheme being put: “The world’s being round” iff the world’s being round. What being the referent of a wentence? My calling it a “state”. World’s being round, being a state.  “The world’s being round” referring to a state. This showing that the English sentence “the world is round” referring to the same state. “The world’s being flat” failing to refer. This showing that the English sentence “The world is flat” failing to refer.  My concluding: truth being reference and falsity being failure of reference.

Advertisements

Written by ilhan inan

March 6, 2012 at 8:57 am

39 Responses

Subscribe to comments with RSS.

  1. I’ve been talking to Ilhan about this for a while.I thought about writing this comment in Wenglish – but then thought better of it.
    I guess the way I understand the basic idea (aand motivation?) is that any sentence in English (“p”) can be translated into a Wenglish noun phrase (“p+”) – and then the question of whether “p” is true can be restated as whether “p+” refers or not. So, if Wenglish has the same expressive power as English, we can give a reductionist account of truth in terms of reference. Ilhan: do you think this is a fair characterisation of your position and motivation?

    For reasons discussed in my earlier post (” Knowledge is not a propositional attitude“) I guess it is easier for Turks than English speakers to get how this translation of English into Wenglish works. In Turkish the grammatical object of a “propositional attitude verb” is a noun phrase, not a that-clause. [So, if you speak Turkish – translate the English sentence into Turkish and then think about how this sentence is made an object of belief.] In English we can only really do this, vaguely, naturally for sentences in the present tense. So we could translate: (1) “Ilhan believe’s that truth is reference” into (2) “Ilhan believes truth’s being reference.” And this in turn can be translated into (3) “Ilhan’s believing truth’s being reference”. This final expression is not a sentence but a noun phrase – and the idea is that (1) is true if (3) refers. It is false if (3) does not refer.

    Anyway – I guess that the basic idea is not too complicated. One big question is whether Wenglish has the same expressive power as English. One might wonder how to translate English sentences that contain logical operators quantifiers and existential claims (and I know Ilhan has been working on this). I think that this can be done. So, for example: We could translate (4) “if the cat is on the mat then an animal is on the mat” into (5) “the following of an animal’s being on the mat from the cat’s being on the mat” and (6) “either he cat is on the mat or a dog is on the mat” into (7) “the disjunction between the cat’s being on the mat and the dog’s being on the mat”. In (7) we might wonder whether the disjunction is really between the cat’s being on the mat and the dog’s being on the mat or between “the dog’s being on the mat” and “the cat’s being on the mat”.

    One question to ask here might be what is it for a noun phase like (5) to refer. this analysis of truth as reference might seem to require a quite baroque ontology.

    I think that this project has a lot of interesting applications. For example, it can be used to provide an interesting analysis of truth (as Ilhan suggests) and also raises interesting connections about logical operators. I’m particularly interested in it’s implications for the ontology of mental states and attitudes. For it suggests a way of dispensing of propositions from our metal ontology – suggesting that all the work done by propositions can be done by concepts.

    Anyway – I have to finish preparing for class now – so this was a bit rushed. But perhaps I’ll continue this comment later.

    Lucas Thorpe

    March 6, 2012 at 1:31 pm

    • Yes Lucas that is a *fair characterization* of my thoughts…thank you. One note about logical connectives: Let p be the proposition that a is F, and q be the proposition that b is G. Now let us say *if p, then q* in Wenglish: *if a’s being F, then b’s being G.* Same with all the other connectives. So I do not need to use *following from* or any such phrase instead of “if..then…”. Wenglish has all the logical connectives, but they connect wentences not sentences. Now you ask what a conditional sentence refers to. I call it a content-state. So a conditional sentence does not directly talk about the external world. “if..then…” turns out to be a reference-shifting operator. Why? Well normally if you say “the world is round” you intend to refer to the following: the world being round. That is a world-state. But when you use the same sentence in a conditional such as “if the world is round, then science has not fooled us”, you don’t wish to talk about the world being round. Rather you wish to talk about the content of that sentence, not its referent. I call that a content-state. You are attributing a property to a content (or a concept if you like) and not the world. Similarly if you say “Sue believes that the world is round” you wish to talk about the same content-state (rather than a world-state) and say something about how Sue relates to that content. Now knowledge may be different. When you say *Sue knows that the world is round” it may be the case that you refer to the world being round and saying something about how Sue is connected to that state or what you prefer to call a fact. This supports your view to some extent; but I am inclined to think propositional knowledge attributions are ambiguous between two separate readings . So I would not go as far as you in saying that knowledge is never propositional.

      ilhan inan

      March 6, 2012 at 7:14 pm

  2. My having written the following on facebook: ‎”Ilhan Inan’s having written an interesting blog post. His construction of an artificial language called Wenglish, the similarity of the being of it to English but without sentences. His appealing to this language being a part of his defence of the claim for truth’s being reference. My having commented.”

    Lucas Thorpe

    March 6, 2012 at 1:48 pm

  3. Ok – here’s my final thought (for now):

    When it comes to logical connectives I guess that a natural way to think about what is being linked are concepts (or perhaps terms]. So Perhaps a better way to translate: ” “if the cat is on the mat then an animal is on the mat” into Wenglish would be “the following of the concept “an animal’s being on the mat” from the concept “the cat’s being on the mat”. (and from our conversations I think you might think something similar – but would express it in less mentalistic terms).

    Now I guess there might be problems with sentences that contain a whole bunch of embedded logical connectives as we might end up with a whole hierarchy of meta-concepts. So for example: How would you translate: “if ( if A is B then A is C) then A is D”?

    One suggestion might be: “the following of the the concept (A’s being D) from the concept [the following of the concept (A’s being C) from the concept (A’s being B)]” – using brackets to indicate names(?) of concepts.

    If one wants to use this language to model mental ontology – this might provide a nice account of why thinking logically is so hard. But I’m curious as to what sort of logic one might end up with. Is it possible to interpret the logical connectives in this language classically? And if not it would seem that this language does not have the same expressive power as English. which would I guess be a problem.

    By the way: as Arman Besler pointed out in their comments to my earlier post – the sort of proposal we’re defending (for slightly different reasons) has similarities to the view Leibniz defends in ‘Generales Inquisitiones’ [translated into English as: “General investigations concerning the analysis of concepts and truths”]

    Lucas Thorpe

    March 6, 2012 at 4:16 pm

  4. Does Wenglish have interrogatives? Imperatives? How do they work?

    Bill W.

    March 6, 2012 at 8:24 pm

    • It definitely does have interrogatives. In fact the idea of Wenglish first emerged as I was trying to give an account of interrogatives rather than declaratives. I hold that curiosity is an intentional mental state in a weak sense: that is in being curious we direct our minds towards an unknown object, though that object may turn out not to exist. A natural consequence of this is that every question asked out of curiosity must involve an attempt to refer to an unknown object. To cut things short every question can in fact be put in terms of a definite description whose referent is being sought. So the Wenglish speakers put a question mark at the end of a definite description to ask questions. “Who is F?” for example can be expressed in Wenglish as “the person who is F?”; “where is x?” translates as “the location of x?”, “why is F?” translates as “the cause of F?” or “the reason for F” etc. When it comes to direct yes/no questions, they just construct a wentence and put a question mark at the end. “The world being round?”, “the existence of the Higgs Boson?”. “Yes” as an answer means that the definite description has a referent (which is what we call “true”) and “no” means that it fails to refer (which is what we call “false”.)

      Imperatives….well I haven’t come that far yet.

      ilhan inan

      March 6, 2012 at 9:28 pm

  5. A shallow question: Why it it called Wenglish? Why W, I mean.

    Lewis, (after Jonathan Swift in _Gulliver’s Travels_) talks about Lagadonian, in which objects are names for themselves and properties are predicates of themselves. Now, Wenglish doesn’t have predicates but descriptional functions, but it seems that functions are logically a lot more like properties than they are like predicates. So, maybe Lagadonian has no predicates but descriptional functions as well. Maybe, all along, Lagadonian had the same features as Wenglish.

    I just saw that recently Benj Hellie put Lagadonian to use in explaining veridical perception. Possibility of your finding it interesting?

    iremkurtsalsteen

    March 6, 2012 at 8:59 pm

    • Hi Irem,

      Derek Ball also appeals to Lagadonian in his paper on Cosmic Hermeneutics. I’m not sure if this is in print yet but he presented it at Bogazici last year. (and the slides are in our departmental drop-box).

      Lucas Thorpe

      March 7, 2012 at 3:07 pm

  6. Facetiously, in this corner of the world, Wenglish is a mixture of Welsh and English. See, for example, http://talktidy.com/

    Hannah Dee

    March 6, 2012 at 9:41 pm

    • Wow I am speechless. Thanks Hannah Dee for letting us know about this. I thought I had simply made up a name, but looks like it is already in use. Maybe following Kripke’s schmidentity example I should call my language “Schmenglish” then. If I say “Denglish” it may turn out to be a Danish dialect of English. Of course best would be to discover that the actual Wenglish dialect of English works similar to how I have described my hypothetical language.

      ilhan inan

      March 6, 2012 at 10:03 pm

  7. Irem that is very interesting about the Lagadonian language, which I will have a look at soon…Meanwhile I think there is a use/mention sloppiness in my post. When I said that rather than predicates Wenglish has descriptional “functions”, I should have really used the phrase “functors” which are linguistic entities just like predicates. So the predicate ” x is white” turns into “the whiteness of x” in Wenglish, or perhaps better is “x being white”. Now that is a functor that refers to a partial function that maps every white object to its state of being white (which you may call a “fact”), and it maps every non-white object to nothing. In this sense they are just like our normal descriptional functions such as *the wife of x*. (I use asterisk to talk about the content.) This is a partial function that maps every married man to his wife and maps every bachelor to nothing. So I am not sure how similar Wenglish is to Lagadonian, but in any case I am curious about it. I think Borges has a story that makes use of a hypothetical language very similar to this.

    ilhan inan

    March 6, 2012 at 9:42 pm

  8. one other curiosity: double negation. not referring of my cat being on the mat not referring is the same as my cat being on the mat referring?

    bernakilinc

    March 7, 2012 at 12:42 am

    • Negation, just like the other logical connectives, is a reference shifting operator. When you say p, normally you wish to refer to the world. But when you negate that and say not-p you wish to say something about p not the world. So if “the cat is not on the mat” is true it does not refer to a negative fact (there are no such things); rather it refers to the content *the cat’s being on the mat* and says of it that it is non-referring. When you negate again you get one more shift of reference and it gets more and more complicated. So the short answer to your question is that p and non-not p do not say the same thing. However they may well be logically equivalent depending on your system of logic. Logical equivalence of course does not mean synonomy.

      ilhan inan

      March 7, 2012 at 11:16 am

  9. Very interesting, thought stimulating stuff İlhan.

    I have several questions below, some (possibly all) of them naive.

    ** I think you would want to say:

    ‘P’ is true iff ‘WP’ refers.

    Refers to what? In the familiar, Tarskian

    ‘P’ is true iff P

    P is a fact. While you are giving up “truth” are you giving up “facts” too. (No pun intended.)

    ** Is your theory an eliminativistic or reductionistic theory about “truth” and about “fact”?

    ** In fuzzy logic truth has degrees. Does reference have degrees too? How would one cash out “degrees of reference”? (I assume you wouldn’t want your theory to rule out fuzzy logic.)

    ** … it maps every non-white object to nothing. … This is a partial function that maps every married man to his wife and maps every bachelor to nothing.

    What is the status of “nothing” in the ontology of WP? In your statement, “to say that a sentence is false is to say that it fails to refer”, what are we to understand by “failure to refer”? Is it the same thing as “referring to nothing”? Is WP’s failing to refer a fact or some kind of an object in the ontology of WEnglish?

    ** How would you express the Liar Sentence “This sentence is false,” and the Truth-Teller sentence “This sentence is true” in WEnglish while preserving the paradoxicality or weirdness of these two sentences?

    Erdinç Sayan

    March 7, 2012 at 4:58 pm

    • Hi Erdinc,

      On the liar’s paradox in Wenglish: How about: “This Wentence’s failing to refer” [with a ‘wentence’ being the Wenglish equivalent of a sentence.]
      Does this wentence refer?

      Lucas Thorpe

      March 7, 2012 at 5:14 pm

      • That’s what came to my mind too. But how do we develop a WLiar Paradox out of it? My failing to see it.

        Erdinç Sayan

        March 7, 2012 at 6:41 pm

      • It seems to me that you get the same paradox, but one advantage is that it brings the liar paradox closer to paradoxes like Berry’s. I have been thinking about this especially after Hartry Field’s talk couple of weeks ago who interestingly modified Berry’ paradox so that he would get rid of the definite description and turn it into a predicate. I rather like the original version given that it is closer to the Wenglish version of the liar.

        ilhan inan

        March 7, 2012 at 7:09 pm

  10. Thanks Erdinc hocam:

    you say:

    ** I think you would want to say:
    ‘P’ is true iff ‘WP’ refers.
    Refers to what? In the familiar, Tarskian
    ‘P’ is true iff P
    P is a fact. While you are giving up “truth” are you giving up “facts” too. (No pun intended.)

    I am not giving up truth at all. All I am saying is that truth is a form of reference. Suppose “Socrates is wise” is a true sentence. Then on my view all this means is that it refers. To what? My claim is that it refers to Socrates’ being wise. I call that a “state”, you may wish to call it a fact. The reason I call it a “state” rather than a fact is because I wish to stay neutral with respect to the realist/antirealist debate as much as possible.

    ** Is your theory an eliminativistic or reductionistic theory about “truth” and about “fact”?

    Given that truth can be reduced to reference, whether I am an eliminativist about truth depends on whether the reference relation can be eliminated. I dont think that it can. So if you press hard I would say I do not subscribe to a deflationist, or a redundancy theory of truth.

    ** In fuzzy logic truth has degrees. Does reference have degrees too? How would one cash out “degrees of reference”? (I assume you wouldn’t want your theory to rule out fuzzy logic.)

    No I wouldnt. Suppose it is 10 degrees C outside and I say that it is 9. What I say is close to being true. Have I referred to the fact of it being 10 degrees outside? I came close. In that sense we may perhaps talk about degrees of reference as well. Vagueness is another problem. In Sorites types of cases there will also be indeterminacy of reference.

    ** … it maps every non-white object to nothing. … This is a partial function that maps every married man to his wife and maps every bachelor to nothing.

    What is the status of “nothing” in the ontology of WP? In your statement, “to say that a sentence is false is to say that it fails to refer”, what are we to understand by “failure to refer”? Is it the same thing as “referring to nothing”? Is WP’s failing to refer a fact or some kind of an object in the ontology of WEnglish?

    By “nothing” here I mean “there is no such thing as…”. Failing to refer is not referring to a non-existing entity, it is simply the case that the term lacks a referent. Just like “the largest prime number” fails to refer, “the earth is flat” fails to refer.

    ** How would you express the Liar Sentence “This sentence is false,” and the Truth-Teller sentence “This sentence is true” in WEnglish while preserving the paradoxicality or weirdness of these two sentences?

    I have been working on that, and it seems that you get the same kind of paradox, though there are certain interesting new twists. If I come up with a new solution you would be the first to know:)

    Thanks Erdinc…I read your new version of the raven paradox today…neat…is this published by the way?

    ilhan inan

    March 7, 2012 at 7:02 pm

    • Thanks İlhan hoca. Nope, none of my posts have been published.

      Erdinç Sayan

      March 8, 2012 at 3:05 am

    • Ilhan’s central claim is this (as I mentioned in my March 7 comment):

      (W) ‘S’ is true iff ‘WS’ refers

      Unfortunately, our concepts of truth and falsehood may not be totally isomorphic to our concepts of referring and failing to refer. Consider:
      “John sleeps glo blu glop lop”
      which is neither true nor false, because it is gibberish. But
      “John’s sleeping glo blu glop lop”
      seems simply to fail to refer to anything, hence the first sentence would have to be false by (W).

      There may arise a similar problem with category mistakes. Consider:
      “Green ideas sleep furiousy”
      which is neither true nor false. But
      “Green ideas’s sleeping furiousy”
      wouldn’t seem to refer to anything, and hence the first sentence is false by (W), rather than neither true nor false. Ilhan might object that the status of “Green ideas sleeping furiousy” is neither-referring-nor-failing-to refer. But even this move (as problematic as it looks) wouldn’t seem to handle the gibberish case.

      Another symptom of nonisomorphism can perhaps be found in the Liar Sentence:
      (L) This sentence is false
      which seems neither true nor false but paradoxical (at least until somebody solves this paradox). The Wenglish version of it appears to be (following Lucas’ suggestion of March 7):
      (WL) This wentence’s failing to refer.
      But (WL) does refer—it is referring to itself when it says “This wentence…”. Hence (L) must be false by (W), rather than paradoxical.

      Similarly for the Truth-Teller Sentence:
      (T) This sentence is true.
      In Wenglish (T) seems to be saying:
      (WT) This wentence’s referring.
      Once again, (WT) does refer, as it is talking about itself. Hence, (T) comes out true by (W), rather than “weird.”

      Trying to fix the problem by translating (L) into Wenglish as follows, for example, wouldn’t seem to work:
      (WL’) This wentence’s referring to its failing to refer.
      For (WL’) is referring—to itself—rather than failing to refer. Therefore its allegation of its referring to its own failure to refer is mistaken. So, if (WL’) is the correct translation into Wenglish of (L), then (L) comes out false and not paradoxical.

      (The situation in (WL’) may be found a little confusing. Let me use the following analogy:
      ‘Atatürk’’s referring to its failing to refer.
      ‘Atatürk’ does refer, so this sentence’s talking about ‘Atatürk’’s referring to ‘Atatürk’’s failure to refer is a case of, as Ilhan might put it, “referring to nothing.”)

      Erdinç Sayan

      March 12, 2012 at 2:49 am

      • Erdinc thanks for all this…ok. here are a few remarks on some of these points:

        My table fails to refer but it is not false. Why not? It is not the right kind of entity. (In some weird language it might be.) For a term to be called “false” it is not sufficient that it fails to refer, it must also fulfill certain conditions for it to qualify. I built all these into the conditions for a term to be a “wentence”. Your gibberish case does not qualify. The category mistake case perhaps doesn’t either. The more problematic cases involve sentences with vague terms and sentences with non-referring terms Concerning the latter: it appears that I am committed to the Russellian idea that “the present king of France is bald” is false (despite the fact that I was a Fregean on this issue for about 15 years.) Similar points came up in different guises when I gave the talk in Virginia and in Milan, so I decided to add a separate section on vagueness and another one on sentences with non-referring terms.

        A point about vagueness: note that if there are cases in which it is simply indeterminate whether a sentence is true or false, then this will translate into Wenglish as indeterminacy in reference. It is not correct to say that such sentences, if they do in fact exist, are neither true nor false. They are simply indeterminate. So it would be wrong to say that such wentences fail to refer. Rather we should really say that it is indeterminate whether they do refer. (Mark and Irem may disagree with this.) The same may apply to the liar sentence, but that requires some extra special treatment anyway, something I have to work on. Perhaps what needs to be shown is that the translation of the liar “sentence” does not qualify as a wentence in Wenglish.

        ….

        Erdinc after thinking about your comments I am now thinking that I should add a new predicate into Wenglish. I might want to call it “walsh” which will correspond to our “false”. Now I can say that any term that is walsh fails to refer, but not any term that fails to refer is walsh.

        Thanks again…

        ilhan inan

        March 13, 2012 at 12:51 pm

        • Thanks Ilhan hocam.

          {{ For a term to be called “false” it is not sufficient that it fails to refer, it must also fulfill certain conditions for it to qualify. I built all these into the conditions for a term to be a “wentence”. Your gibberish case does not qualify. The category mistake case perhaps doesn’t either. }}

          I thought we would translate the sentences
          “John sleeps glo blu glop lop”
          and
          “Green ideas sleep furiousy”
          into Wenglish as
          “John’s sleeping glo blu glop lop”
          and
          “Green ideas’s sleeping furiousy,”
          respectively. These wentences lack reference, and consequently the original English sentences erroneously come out false via what I labelled (W). But you want to translate the original sentences into Wenglish not in way I did, but in such a way that the resulting Wenglish sentences are indeterminate as to whether they refer or not. Could you give us a sense of how the wentences corresponding to these English sentences go?

          Erdinç Sayan

          March 14, 2012 at 12:57 am

      • Erdinc hocam I doubt that the phrase “John sleeps glo blu glop lop” qualifies as a sentence in English. So its translation into Wenglish does not even come up as an issue. (This is on the assumption that you cannot translate gibberish; an assumption that might be given up in certain cases.) Now “Green ideas sleep furiously” might qualify as a sentence of English. If so its Wenglish translation would be something like “the furiously sleeping of green ideas” which fails to refer of course. But is it “walsh” (or false)? Well that depends on whether it qualifies as a wentence. If it doesn’t then its failure to refer would not be sufficient to say that it is walsh. If it is a wentence then I would say that it is walsh. I can live with that. Now of course I have not told you that some more intellectually sophisticated Wenglish philosophers distinguish between two kinds of walshities: they distinguish between say “the world being flat” being walsh, and “the present king of France being bald” being walsh. Your example (or Chomsky’s) would belong to the latter group I guess because it has a non-referring part (i.e. “green ideas”).

        ilhan inan

        March 14, 2012 at 4:13 pm

        • {{ I doubt that the phrase “John sleeps glo blu glop lop” qualifies as a sentence in English. So its translation into Wenglish does not even come up as an issue. (This is on the assumption that you cannot translate gibberish; an assumption that might be given up in certain cases.) Now “Green ideas sleep furiously” might qualify as a sentence of English. If so its Wenglish translation would be something like “the furiously sleeping of green ideas” which fails to refer of course. But is it “walsh” (or false)? Well that depends on whether it qualifies as a wentence. If it doesn’t then its failure to refer would not be sufficient to say that it is walsh. If it is a wentence then I would say that it is walsh. I can live with that. }}

          Sorry to bother you with more comments Ilhan hoca. But it’s your fault to give us such a thought-provoking theory…

          I presumed that for Wenglish to have the same expressive power as English, not only questions, imperatives, and declarative sentences must be translatable into Wenglish, but also category mistakes (assuming there be such things) must be translatable as wategory mistakes (and NOT as wentences that are walsh), paradoxes in English must be translatable as waradoxes and so on. (OK, I guess we don’t have to insist that “gibberish sentences” has to be translatable into “wibberish wentences.”)

          Erdinç Sayan

          March 14, 2012 at 11:34 pm

        • A new edition of my previous comment:

          Sorry to bother you with edited comments Ilhan hoca. But it’s your fault to give us such a thought-provoking theory…

          I presumed that for Wenglish to have the same expressive power as English, not only questions, imperatives, and declarative sentences must be translatable into Wenglish, but also category mistakes (assuming there be such things) must be translatable as wategory mistakes (and NOT as wentences that are walsh), paradoxes in English must be translatable as waradoxes and so on. (OK, I guess we don’t have to insist that “gibberish sentences” has to be translatable into “wibberish wentences.”) No?

          Erdinç Sayan

          March 18, 2012 at 2:13 am

        • Erdinc really sorry I missed your last comment. About category errors: “the furiously sleeping of green ideas” commits a category mistake in Wenglish as much as its English counterpart does so in English. A category mistake in Wenglish is not one that is committed by applying a predicate to the wrong kind of entity; there are no predicates in Wenglish. Rather it is substituting the wrong kind of term in an argument place of a descriptional functor. You have “the furiously F-ing of x”, and it works if you substitute “running” for F, and “Socrates” for x to get “the furiously running of Socrates”. But if we place “sleeping” there we get a category mistake. Now depending how you deal with this in English you would seem to have a similar way of dealing with it in Wenglish. If you say “Green ideas sleep furiously” is a sentence of English that is neither true nor false, then I do admit we have a problem. The corresponding term “the furiously sleeping of green ideas” would then have to be a wentence in Wenglish which fails to refer. But that would imply that it is walse (I called it “walsh” before, but I guess “walse” is better). So then the English sentence should really be false as well. This is the problem you have in mind, right? And my initial response to that is to claim that the expression “the furiously sleeping of green ideas” is not a wentence of Wenglish. So then it should follow that “Green ideas sleep furiously” is not a sentence of English either. If you insist that it is a sentence, then I would be forced to conclude that it is false. (I am inclined to think that it is not a sentence, because it does not “say” anything in some sense of “say”.) So what really needs to be done is to give a theory of sentencehood (wentencehood) so that such expressions do not qualify as sentences.

          About the liar paradox…I follow all that you say, but it requires a lengthier discussion (and a bit more thinking on my part) which I will save for later. Thanks!

          Hope we get a chance get together some time sevgili Erdinc hocam…

          ilhan inan

          March 18, 2012 at 6:15 am

  11. Ilhan, I wondered what happens to the de re/de dicto distinction in Wenglish. I worked on translating “The number of planets is necessarily even”, which is true (with Pluto gone), as opposed to, “Necessarily, the number of planets is even”.
    De dicto reading: The necessity of reference for “the being even of the number of planets”
    De re reading: The necessity of the being even of the number of planets T

    Problem is, I can’t see what the AMBIGUOUS Wenglish sentence would be. This may be a good thing for Wenglish, of course.

    iremkurtsalsteen

    March 7, 2012 at 7:55 pm

    • Irem, I think that the ambiguity is still there in Wenglish. Here is a “natural” way to say it:

      De Dicto: Necessarily, the number of planets’ being even.
      De Re: The number of planets’ being necessarily even.

      The first one is false because “the number of planets’ being even” refers to different facts in different possible worlds where that number is a different even number (2,4,6,..) and refers to nothing in possible worlds where that number is not even. The second one is true because 8 is even in all possible worlds, in other words 8’s being even is a fact of every possible world (in which 8 exists). There are some interesting issues that relate to this concerning how to account for rigidity in Wenglish. I will talk about that in more detail in the seminar.

      ilhan inan

      March 8, 2012 at 8:37 am

  12. Hi Ilhan,

    You didn’t really answer my previous question about logic.
    Here’s another way of putting it: How do you explain modus ponens?
    In English there is an intuitive way of thinking of how modus ponens works:

    we have:
    p –> q
    p
    therefore
    q

    And we can think of p–>q as something like a function and p as something like an argument that can be plugged into this.

    But in Wenglish, where logical operators are reference shifting we have something like:

    “p” –> “q”
    p
    Therefore
    q

    So, what justifies plugging p into “p” –> “q”?

    I think there will also be a problem explaining logical equivalence if you take the logical operators to shift reference.

    Lucas Thorpe

    March 8, 2012 at 12:10 pm

    • Lucas this really is a difficult matter and I certainly do not have a fully worked out solution. (Initially I thought I came close to it but then I realized a flaw in it, something like an infinite regress problem.) Here are a few thoughts: the claim that some of the logical connectives are reference shifting operators should really be treated as a separate claim. I believe there are good arguments in favor of it even if I am wrong that truth is a form of reference. When we assert if p then q, we assert neither p nor q. That really tells us that we do not intend to refer to their customary referents. And if not, we should be referring to their contents. If you take the traditional view and take those contents to be propositions, then what we are really saying is that proposition q in some sense follows from proposition p. After that if you assert p, you infer q and assert it. One may just take that to be the primitive MP rule that requires no further formal justification. There are however independent reasons for saying “if…then” is a reference shifting operator. If falsity is simply a failure of reference as I claim, then how could a conditional be true when a part of it is false and thus lacking a referent?. If you do not wish to give up compositionality for reference you have to deal with this. The other issue of course is logical equivalence as you say, and that really is the other side of the coin. Here again we shall have to take some of the inference rules as being primitive. Hope we talk about these issues some time.

      ilhan inan

      March 9, 2012 at 8:56 am

      • Hi Ilhan,

        I can understand why you think negation is a reference shifting operator. As a natural way to understand negation in the language is in terms of “A’s being B” not referring.

        But I guess that maybe you’re going to need a non reference shifting conditional. I guess if Wenglish is to have the same expressive power as English (which I guess you need for your reductionist project) you should be able to express the classical connectives in Wenglish. And at some point I think you’re going to need to be able to say:

        p → “p”

        And you can’t do this if the connectives necessarily shift the reference. But I don’t see why intuitively we have to assume that the conditional is reference shifting in Wenglish. (I guess that you only require a shift in reference if the connective is understood in truth functional terms – so one could have a non-truth functional non-reference shifting conditional)

        I also think there might be a non reference shifting negation. So a natural way of reading not-“A\s being B” is as “A’s not-being B”. And so the negation understood in these terms might be understood in terms of a permission to write something like the dual of the wentence. This won’t give you classical negation. I guess it would be a non-truth functional negation.

        Lucas Thorpe

        March 9, 2012 at 3:05 pm

  13. “Queen Victoria was crowned” is true in English, but I don’t see how it can be expressed by a Wenglish referring expression, since it doesn’t have a unique singular truth condition. Queen Victoria was crowned first as Queen of England and later as Empress of India but that doesn’t affect the truth of “Queen Victoria was crowned”, whereas “The (past) crowning of Queen Victoria” does require a unique referent.

    Karl Pfeifer

    March 8, 2012 at 6:36 pm

    • That is a very interesting example indeed, thanks. I would say the more natural translation of your sentence into Wenglish would be: “Queen Victoria’s having been crowned (in the past)”. This should not be taken to be an attempt to make reference to a *crowning event*. If it did then we would have a problem, given that there are two such events. Rather it makes reference to a person’s having a certain property. The property is *having been crowned* but the sentence does not say anything about what event or events enabled that person to acquire that property. So there are two events, but there is only one property, and that is what matters. Does that resolve the problem?

      ilhan inan

      March 9, 2012 at 9:10 am

      • I don’t think so. Let me try a different example without the complications of tense and occurrences over time:

        S =df “A wasp is buzzing overhead”

        S can be true if any nonzero number of wasps is/are buzzing overhead. No particular wasp is referred to in S, so S doesn’t have a have a unique singular satisfaction condition that can be correlated with the referent of a Wenglish referring expression like “the wasp’s buzzing overhead”, whereas “a wasp’s buzzing overhead” is an indefinite description and therefore ostensibly nonreferring.

        Karl Pfeifer

        March 13, 2012 at 9:54 pm

        • Hi Karl, if you don’t take an indefinite description to be a referring expression, then you could give it a Russellian analysis: “A wasp is buzzing overhead” is then an existential claim: “There is an x, x is a wasp and x is buzzing overhead.” Such a sentence then may be taken to refer to a concept and say of that concept that it is instantiated. The concept in question could simply be put as *being a wasp buzzing overhead*. In Wenglish you could say what you say as “There being a being a wasp buzzing overhead”. I take that to involve reference to a concept, or better to call it a “content”; and then we say of it that it is instantiated. That is why Frege and Russell called *existence* a second-order or second-level property. The referent of that wentence I like to call a content-state; the state of a certain content being instantiated. So if there is no reference to a particular wasp in the sentence as you say, then there must be reference to a particular content (concept), and the sentence says that it has the property of being instantiated. That is how I deal with existential claims and some sentences with indefinite descriptions. Is that satisfactory?

          ilhan inan

          March 18, 2012 at 12:33 pm

  14. Interesting post. What do you say about cases in English where we apparenty use state-descriptions for states of affairs that to not obtain?

    For example: suppose John loves Mary but Mary will never love him. It seems we might truly say ‘Mary’s loving John would make the latter very happy’. This seems to be acceptable English, but I cannot see how you would translate it properly into Wenglish – i.e., so that it doesn’t imply that Mary loves John.

    It seems here you need a reference shift from states to content-states (as you call them). You could perhaps say that the above is a disguised counterfactual conditional, and then deploy your reference-shifting story about connectives.

    But what, then, about the sentence ‘Mary’s loving John is desired by John’? Here, shifting to content-states seems implausible: John does not desire some meaning-like entity, he desires Mary’s actually loving him.

    Tristan Haze

    March 13, 2012 at 8:02 am

    • In English you could say “John desires that Mary loves John”. The content of John’s desire is a proposition. So if we follow Frege here, there is reference to a certain John and there is also reference to a proposition; and we say something about how the two are related. It is exactly the same in Wenglish: “John’s desiring of Marry’s loving John”. Here the embedded singular term “Marry’s loving John” does not refer to a world state but rather it refers to its content. That does not imply that John desires a content, whatever that may mean. Rather the verb “to desire” in the wentence relates this content to John. *Desiring* is having a certain attitude towards that content. So if you take *desire* to be a propositional attitude in English, it is very similar in Wenglish (though it would be a “wropositional attitude”).

      ilhan inan

      March 14, 2012 at 4:28 pm

  15. {{ If you say “Green ideas sleep furiously” is a sentence of English that is neither true nor false, then I do admit we have a problem. The corresponding term “the furiously sleeping of green ideas” would then have to be a wentence in Wenglish which fails to refer. But that would imply that it is walse (I called it “walsh” before, but I guess “walse” is better). So then the English sentence should really be false as well. This is the problem you have in mind, right? }}

    Precisely.

    {{ And my initial response to that is to claim that the expression “the furiously sleeping of green ideas” is not a wentence of Wenglish. So then it should follow that “Green ideas sleep furiously” is not a sentence of English either. If you insist that it is a sentence, }}

    I wouldn’t.

    {{ then I would be forced to conclude that it is false. (I am inclined to think that it is not a sentence, because it does not “say” anything in some sense of “say”.) So what really needs to be done is to give a theory of sentencehood (wentencehood) so that such expressions do not qualify as sentences. }}

    Good call.

    {{ About the liar paradox…I follow all that you say, but it requires a lengthier discussion (and a bit more thinking on my part) which I will save for later. Thanks! }}

    You are most welcome. I thank you. I will be interested to know how you deal with waradoxes. Perhaps the Liar Paradox is not the only paradox in English you will have to “waradoxize.” (See, I can invent words too…)

    {{ Hope we get a chance get together some time sevgili Erdinc hocam… }}

    Insallah sevgili Ilhan hoca.

    Erdinç Sayan

    March 18, 2012 at 1:39 pm

  16. Wow. I think it’s possible that I think in Wenglish!

    For this reason, I can’t reason very well propositionally about it. And so don’t have anything to add to my “Wow.” Which I would nevertheless like to reiterate. Except for “Not seeing reference as exhausting the work done by ‘true.'”

    The Nous Bros.

    April 7, 2012 at 1:51 am


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: