[hist-analytic] Deflationary Truth and Syncategoremata

Baynesr at comcast.net Baynesr at comcast.net
Tue Dec 28 11:54:39 EST 2010

By 'meaning in isolation' I did not mean a word in isolation. I meant that a word in a syncategorematic position will contribute no meaning to the whole, as it would contribute to the meaning of a sentence were it to occur in that sentence outside of a syncategorematic position. Take Quine's example: 'nine' in 'canine'. A word like 'poor' in 'poor violinist' where the meaning is that the guy can't play the violin well has, on the view under discussion, the same lack of contribution to the whole as 'nine' in 'canine'. So the issue I was raising wasn't about words in isolation in your sense. 

More appropriate to the sense(s) you indicate is the issue of "lexical syntax," that is, a syntax based on syntactic, and possibly semantical features of individual entries in the lexicon. Here there is an issue of words in isolation. The deep underlying semantical, not syntactical, issue is the notion of compositionality: whether the meaning of a syntactic complex is determined as a "sum" of the meanings of the components in the syntactic construction at issue. This is in my opinion one of the main problems with a lot of model theoretic linguistics. 



----- Original Message ----- 
From: "Scott Holbrook" <scott.holbrook at gmail.com> 
To: Baynesr at comcast.net, "Landspeedrecord" <landspeedrecord at gmail.com> 
Cc: hist-analytic at simplelists.co.uk 
Sent: Monday, December 27, 2010 7:26:55 AM 
Subject: Re: Deflationary Truth and Syncategoremata 

This is probably moving away from deflationary-truth, but... 

I'm somewhat unclear about what constitutes an "isolated word" (if such a thing is even possible) and how syntax (which I take to deal with the merely the arrangement of symbols, in the case of sentences the syntax involves words and puncuation) could be relevant for a "word in isolation." Or, how an "isolated word" can even have a syntactical position... 

I can think of three ways, off-hand, to "isolate" a word. None of them get very far... 

1) Perhaps some alien spacecraft is discovered and only a single "a string of letters arranged by concatenation" remained in some obscure, well protected, place on the hull (all other markings have eroded). So we have, supposedly, a single word from an alien language. And here, of course, the obvious question is why we should count this string of symbols as a word? (And a less obvious question, can we imagine that our artwork could be mistaken for words by an alien...that walls of art were sentences, galleries were chapters and museums were novels?) I just don't see how, without something else , we can legitimately say that this symbolic conglomerate is a word (other than ex hypothesi). Sadly, it seems that once I bring in the something else, that the isolation has been lifted. (note: Knowing that they have language, (which an interstellar craft suggests they do) is not the same as knowing what constitutes the language (e.g., the markings may be the result of an industrial process used to make the crafts)...another example, the aliens in Avatar had no written language, nor a need for one...strictly speaking, they didn't need to use any meduim of communication accessible to human senses.) 

2) I'm driving through the Mohavi Desert, I happen to glance up and the letters "CAT" are on a billboard, in the middle of nowhere. Now, if I don't understand English, I may presume that "CAT" is a word (or they may presume it's some crazy American artwork). If I do know English, then I'm certain that "CAT" is (or could be) a word. The question, however, is whether or not the wordhoodness of "CAT" was determined in isolation...and I don't think that it was. In the former case, there's really no way to claim that "CAT" is a word (anymore than "aDOJFGHDAofa" is a word). In the later case, authority for identifying this word comes from an indefinite amount of other words than CAN surround it. That is, we don't recognize wordhoodness as being some intrinsic feature of "CAT," we remove it from it's "isolation" and realize (in a flash, for native speakers) a host of possible "syntactic fits" lots of which have meaning but some that don't. A related issue is knowing that the string of letters on a particualr sign is the name of town, having never known a town of that name existed. 

3) An off the wall interjection, or reply, in a conversation. For example 
A: Hello. 
B: No. 
A: Are you alright? 
B: Banana. 
A: I'm going to go now. 
B: Visions of sugarplums. 

All of B's lines have their places, those places just aren't here. There are two paths. First, one may say that in this particular instance, B's utterances are meaningless, that is they are not words...in this instance. But they would be words if uttered in a more appropriate context. Of course, if they are not words in this context, then this is not an example of an isolated word. The second option, is to say that yes, these are words, but that B is just misusing them. Perhaps he's not a proficient English user. They are "out of place" so to speak, but still carry meaning. Well, that's fine, but "out of place" or "being misused" can't be predicated of isolated things. (Imagine giving a "word association test" to someone without telling them what your doing...unless the person has superhuman patience, you're not going to get very far.) 

I think these are three distinct senses of isolation. I'd like to know which, if any, is the isolation spoken of in the post: 

In the first, there is a total isolation from anything having to do with the symbols (in which case it's not clear how to determine if a thing is a word or not). 

In the second case, there is a linguistic isolation (maybe a bad name for it) where the "word" is given without other words around it, yet access to the 'setting' of the word is given (e.g., the name of the unknown town on the sign). 

The third is, I guess, a semantic isolation, of sorts. The "words" are used in such a way that they don't have any semantic import. They have been isolated from their conditions of being meaningful (once again, maybe this is a poor choice of wording). That is, we'd find it impossible to carry on a conversation with B. We'd have no way of knowing what, or if, he means and no reason to believe that he knows what, or if, we mean. Of course, if we don't know what, or if he means, then we don't know, without knowing something else, if he is uttering words (babies make a lot of utterances most of which are not words). 

So, can we just pick some word at random from English...Sure. Can we talk about whether that word would mean something all by itself (in isolation, that is) after we have picked it...I want to say, not really. Sure, we could talk about all the possible meanings that string of characters may have, should it be injected back in the language, and even the ones it had before it was extracted. We can also imagine all the questions to which this single string would have been an acceptable response or all the times we may been moved to utter this word by itself. But, I really want to resist calling it a word at this point. I could just as easily have made up a random set of characters (e.g. asdfglkjh) and then started assigning uses. 

It just doesn't seem as if we can understand words, as words , in isolation. I almost want to say that (except in some philosophically trivial instances), "isolated word" is an oxymoron. 



On Fri, Dec 24, 2010 at 7:52 PM, < Baynesr at comcast.net > wrote: 

"So, if truth resides outside us as a property of reality then deflation cannot hold..." 

Yes, this is an often stated view. The issue becomes one of realism vs. verificationism, at some point. If deflation holds then the correspondence theory of truth may be in trouble. But not everyone is a realist...so the debate goes on. Many, e.g. N. Goodman, deny there is a reality. The "reality" depends on your "model." Roger might agree. Don't know. 

The issue of syncategoremata is a syntactical one, seldom semantical. Since the idea came up in Occam some years ago, the idea has been (Quine) that whether a word has a meaning in isolation depends on its syntactical position. In other words, where in our example 'poor' occurs in syncategorematic position it contributes no independent meaning. In fact the radical view is that it is to be viewed as a string of letters arranged by concatenation and nothing else. So 'poor' in 'poor violinist' when it occurs syncategorematically has NO meaning. I think this is basically right, but the concept was used to excess; a worn out toy, so to speak. Like dispositions, in my opinion. 


----- Original Message ----- 
From: "Landspeedrecord" < landspeedrecord at gmail.com > 
To: Baynesr at comcast.net 

Cc: hist-analytic at simplelists.co.uk 
Sent: Thursday, December 23, 2010 7:49:55 PM 
Subject: Re: Deflationary Truth and Syncategoremata 

I wasn't saying your remarks were senseless! I was saying that deflationary theories of truth literally have no sense to me in that there is no conceptual framework I can wrap my brain around that does not substitute the idea of deflation with that of suspension. In other words, the only way I can make sense of deflationary theories is to think of them as being "suspensionary" theories of truth, in which case it all seems silly to me. Rather than challenging any notions, I was more admitting a SERIOUS cognitive gap on my part hoping that someone will clue me in on what I am not getting about deflationary theories of truth. 

As for the word 'poor', in my prior email I was always using it to mean 'lacking wealth', so if you reread my prior email with that in mind, perhaps you might get a different spin on what I was saying? But perhaps it is the other way around, perhaps I am misunderstanding you... I was also assuming you were not jumping between two different meanings of the word 'poor'. If you were, please let me know. I assumed you were always using it to mean impoverished. 

At any rate, I wasn't attempting to muddy things by playing on the ambiguity of the term 'poor'. So for instance, 'poor Arnold' is meant as 'Arnold lacks money', just as I took you to mean that 'poor violinist' meant 'there exists a person and this person plays the violin and also lacks money'. Lacking money isn't modifying the ability to play the violin and the ability to play the violin isn't modifying the lack of money - hence my comment about 'poor' and 'violinist' referring in parallel but not to each other. Whereas with 'poor person' the lack of money IS modifying the proxy placeholder that we cognitively create for the unspecified person, and unlike 'poor violinist' the modification is to a word within the sentence - via the proxy status of the word 'person'; contrast this with 'poor violinist' where the cognitive modification is to an object outside the sentence. I should mention that I am not speaking of things here grammatically, obviously in a grammatical sense the adjective 'poor' modifies the noun 'violinist'. To me, 'violinist' is really an adjective noun combination in disguise: Cognitively, it unpacks to something like 'violin-playing person'. 

'poor violinist' -> 'wealth-lacking violin-playing person' 
'poor person' = 'poor person'. In this case there is nothing to unpack as the cognitive proxy is already part of the sentence. 

Perhaps I am devolving into an implicit Russellian stance unknowingly here (the king of france is bald... etc...) but I don't think so. My point isn't that utterances need to be properly unpacked to determine truth values, what I was saying was that the analogy between 'poor violinist' and 'true sentence' was a poor one due to the implicit way that we treat descriptive nouns like 'violinist', and that if 'poor person' was used instead, then the comparison with 'true sentence' becomes more apt - 'sentence' and 'person' being closer in relation to one another in that they are empty cognitive vessels that we fill with specifics, whereas 'violinist' already comes with a specific attached. 

So... ''S' is true' says nothing more than merely asserting 'S'" - this is where I lose the deflationary viewpoint. The proposition 'S', until specifics are provided, is an empty cognitive placeholder. As such, one of the specifics that we can attach to 'S' is that it is true. This allows us to speak about the structure of 'S' before it is ever asserted/uttered/specified and as such there must be something going on which a deflationary theory cannot account for. Therefore it makes sense to view truth as a property of a proposition, at least relative to the knower. And if this stipulation is necessary to account for truth (as some may know 'S' is true and other may not even know what 'S' is, much less whether it is true or not) then truth must be an actual property of our relation to the proposition, and thus standing objectively OUTSIDE us, if not a property residing in the proposition itself. So, if truth resides outside us as a property of reality then deflation cannot hold, despite a deflationists appeal to the idea that the notion of deflation is not supposed to apply to "meta" propositions of the form ''S' is true'. 

I have a suspicion that I am entirely missing the intent of your initial post, if so, please set me straight! 

On Sun, Dec 12, 2010 at 2:41 PM, < Baynesr at comcast.net > wrote: 

Thanks, Landspeedrecord (?), whoever you are! Much appreciated. Let me just add a clarifying remark and a question or two. We can always go back to some of the details. First, on "syncategorematic." 

X can be a poor violinist without being poor; so Quine's thinking here is that 'x is a poor violinist' does not imply that 'x is poor' on the syncategorematic reading. However, on the categorical reading 'x is a poor violinist' implies that x is both poor and a violinist. He stands in front of the Harvard Coop playing Mozart for quarters. When I used 'modify' my intent was to restrict this to the case where 'poor' in 'poor violinist' modifies x; that is, attributes to the violinist the property of being poor (impoverished) although, maybe, a terrific violinist. My reading of 'poor violinist' on the syncategorematic reading is that 'poor' cannot stand alone as modifying that to which 'poor violinist' is attributed; whereas, in the categorematic reading 'poor' can stand alone to "modify" the "subject" viz. 'violinist', which in such cases can always stand alone. I can't get a sense of 'poor' in your example, 'poor Arnold', of the same sort that I can get in 'Arnold was a poor violinist'. 

My remarks on truth may be senseless. But I'm not sure that the deflationary theory is senseless. It is to say that ''S' is true' says nothing more than would an assertion of 'S'. Now I'm not prepared to either accept or reject this theory, but it does seem to me to make sense. 

Best wishes 


----- Original Message ----- 
From: "Landspeedrecord" < landspeedrecord at gmail.com > 
To: Baynesr at comcast.net , hist-analytic at simplelists.co.uk 
Sent: Sunday, December 12, 2010 12:49:51 PM 
Subject: Re: Deflationary Truth and Syncategoremata 

I don't understand this. I don't understand deflationary theories of truth in general and this cuts nicely to the heart of the matter for me. 

'Poor', in the "syncategorematic" sense is modifying, not another word in the sentence but the referent. So is violinist. They refer in parallel to the referent but not to each other. 

On the other hand, 'poor Arnold' is different. Here 'Arnold' is a proxy/stand-in of the referent - and I am not speaking of rigid designation here. 'Arnold' stands FOR the (hypothetical) person who we refer to as "Arnold". So 'poor' IS modifying 'Arnold'. The property of being poor is added to the mental list (temporarily at least) we are keeping about the properties of 'Arnold'. This is particularly clear in the case where it is understood that there is NO person being referred to, where the name is just being used demonstratively, as it is by me in this case. 

So 'true sentence' is nothing like 'poor violinist' - a much better analogy is 'poor person' or 'poor human'. That is to say, poor is a putative property of the putative person, and true is a putative property of the putative sentence. Until the actual sentence is provided, then yes, in a trivial sense, the term 'truth' is just an emphatic declaration, truth has been deflated - but really what is going on is that truth isn't being deflated... it is being suspended - as is 'poor' when it modifies 'human' - once the referent is given the suspension is OVER. And hence deflation of truth is merely a disguised form of suspended/imaginary truth. 

And so deflationary theories make literally NO sense to me, as I can never get past this issue. 

On Fri, Dec 10, 2010 at 2:33 PM, < Baynesr at comcast.net > wrote: 

Recall Quine's example of syncategoremata: 

'poor violinist' 

Where 'poor' does not modify 'violinist', i.e. in the sense that the violinist may be wealthy. We 

may suppose 'poor violinist' is predicated of 'Jack'. 

Now consider 

'true sentence' 

where 'true' does not modify 'sentence', and where we suppose the true sentence is 'p' (cf. 


So we have it that we might assert ambiguously, either: 

'The violinist is poor' 

'The sentence is true' 

A deflationary theory of truth is consistent with the syncategorematic reading of 'true' in 

'true sentence'. We may assert 

'p is true 

and no more commit ourselves to property attribution that we do when predicating 'poor' of 

the 'violinist' when we say, 

The violinist is poor. 


STeve Bayne 

Pentabarf #5: A Discordian is Prohibited from Believing What he reads. 

Pentabarf #5: A Discordian is Prohibited from Believing What he reads. 

"Conventional people are roused to fury by departures from convention, largely because they regard such departures as a criticism of themselves." 
-- Bertrand Russell 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rbjones.com/pipermail/hist-analytic_rbjones.com/attachments/20101228/bc43a379/attachment-0002.html>

More information about the hist-analytic mailing list