Language, Philosophy of – Page 2 – Maverick Philosopher

The Third Way of the Ostrich

In a comment, the Ostrich writes,

Some early analytic types, including Russell, tried to analyse proper names as disguised descriptions, but Kripke put a lid on that. Thus, on what Devitt calls the Semantic Presupposition, namely that there are no other possible candidates for a name’s meaning other than a descriptive meaning, or the bearer of the name itself, the mainstream analytic position is that the meaning of a proper name is the bearer of the name. The target of Reference and Identity is the Semantic Presupposition.

So far, so good. I agree that with respect to proper names, demonstratives, and indexicals, both description theories and direct reference theories fail. So it makes sense to investigate whether the Semantic Presupposition is a false alternative. But the Third Way of the Ostrich raises questions of its own and they incline me to think that it too leads to an impasse and is in the end No Way, a-poria.

Consider the proper name, 'Moses.' It does not refer to the expression 'the man who led the Israelites out of Egypt.' It refers to a man, not an expression. (9) Thus "'Moses' refers to a man" is true. But what makes it true? One might think that it is true in virtue of a relation that connects the name to a particular man, and thus to something extra-linguistic. But the Ostrich denies that there is an "external reference relation" that relates the name to something extra-linguistic. (9) What makes true the reference statement — "'Moses' refers to a man" — is "an internal relation between the reference statement and some textual or uttered antecedent." (9) It is not clear what this means since it is not clear how the reference statement can have an antecedent. I know what the antecedent of a pronoun is, but what is the antecedent of a sentence or statement? I also know that a statement can be the antecedent of a term. For example, "Snow is white. This everyone agrees to." In this example, the demonstrative 'this' has a statement as an antecedent. What I don't understand is how a statement can have an antecedent. But let that pass.

It is clear what the Ostrich wants to say: there is reference but all reference is intra-linguistic. That contrasts with what I am inclined to say, namely, that while some reference is intra-linguistic, not all reference is. The reference of 'he' is parasitic on the reference of 'Tom' in 'Tom enjoyed the massage he received' and so there is a sense in which the reference of 'he' is intra-linguistic; but 'Tom,' if it refers at all, refers extra-linguistically. In which precise sense is the reference of 'he' in our sample sentence intra-linguistic? Surely the pronoun 'he' does not refer to the name 'Tom'; the pronoun refers to the same item to which 'Tom' refers. So to say that the reference of 'he' is intra-linguistic is just to say that it picks up the reference of its antecedent and would not refer otherwise. Pronoun and noun are co-referential which is to say that they refer to the same item if they refer to anything. But the burden of objective reference is shouldered by the noun, not the pronoun. Or so say I.

The Ostrich's idea here is that "the semantic value of a proper name consists SOLELY in its anaphoric co-reference with its antecedents in a chain of co-referring terms . . . ." (8, my emphasis)* Interpreting, one could say that reference is constituted by co-reference which is always an intra-linguistic matter. This would seem to issue in an objectionable linguistic idealism.

'Asmodeus,' we are told, refers to Asmodeus, so the name refers to something. It refers to a demon, not an expression, similarly as 'Moses' refers to a man, not an expression. But from the fact that 'Asmodeus' refers to something it does not follow that something exists which is the referent of 'Asmodeus.' (10) That is surely true. But it is also true that from the fact that 'Asmodeus' refers to something it does not follow that nothing exists which is the referent of 'Asmodeus.' So the referent of 'Asmodeus' may or may not exist.

I now put the question to the Ostrich: what is it for the referent to exist? We are assuming that there is no such demon as Asmodeus. And yet 'Asmodeus' refers to something. There is a difference between referring to something that does not exist and not referring to anything. Now the Ostrich told us that 'Asmodeus' refers to something. But then something is such that it does not exist, and we are in Meinongian precincts — which is precisely where an ostrich will not stray if he can help it.

So the Ostrich cannot mean that 'Asmodeus' refers to something that does not exist; he must mean that 'Asmodeus' is an empty/vacuous name, i.e., one that does not refer at all, one without a referent. Again, there is a plain difference between a term's having a non-existing referent and a term's having no referent at all.

The trouble with saying that 'Asmodeus' is an empty name, however, is that it conflicts with his theory according to which "the semantic value of a proper name consists SOLELY in its anaphoric co-reference with its antecedents in a chain of co-referring terms . . . ." (8, my emphasis)* There is a conflict with the theory because 'Asmodeus' is a member of a chain of co-referring terms, which implies that 'Asmodeus' has a semantic value, an object, an object which exists simply in virtue of being an object. So Asmodeus exists after all.

The demon cannot both exist and not exist. One might say that that the demon does not exist in reality (outside language) but that it does exist in a language-immanent, 'internal' way as an object constituted by "its anaphoric co-reference with its antecedents in a chain of co-referring terms . . . ." But if the demon does not exist in reality, then Moses does, in which case the reference statement — "'Moses' refers to a man" — must have an external reference relation as part of its truth maker.

If that is denied and reference is intra-linguistic only, then how account for the difference between the existent Moses and the nonexistent Asmodeus? After all, both names belong to chains of co-referring terms. Each name belongs to a narrative.

Is our Ostrich a POMO bird in the end?

__________________________

*I suspect that the Ostrich is using 'semantic value' in the way Gareth Evans uses it, namely, as equivalent to Frege's Bedeutung. Accordingly, the semantic value of a proper name is an object, that of a concept-word (Begriffswort) is a function, and that of a sentence (Satz) is a truth value (Wahrheitswert).

Could Scollay Square be a Meinongian Nonexistent Object?

Bill, newly arrived in Boston, believes falsely that Scollay Square exists and he wants to visit it. Bill asks Kathleen where it is. Kathleen tells him truly that it no longer exists, and Bill believes her. Both use 'Scollay Square' to refer to the same thing, a physical place, one that does not exist. To exist is to exist in reality. 'In reality' means outside the mind; it does not mean in the physical world.

So both Bill and Kathleen use 'Scollay Square' to refer to a physical place that does not exist. The two are not using (tokens of) 'Scollay Square' to refer to Fregean senses or to any similar abstract/ideal item.* Scollay Square is not such an item. It is concrete, i.e., causally active/passive. After all, it was demolished.

Now it could be that reference is routed through sense as Frege maintained. Perhaps there is no road to Bedeutung except through Sinn. Whether or not that is so, when Bill and Kathleen think and talk about Scollay Square, they are not thinking and talking about an abstract object that mediates reference, whether it be thinking reference or linguistic reference. They are thinking and talking about a concrete, physical thing that does not exist.

We also note that Bill and Kathleen are not thinking or talking about anything immanent to consciousness such as a mental content or a mental act. They are referring to a transcendent physical thing that does not exist. Scollay Square is not in the head or in the mind; if it were, it would exist! If memory serves, it was the illustrious Kasimir Twardowski who first made this point, leastways, the first in the post-Brentano discussion.

Therefore, some transcendent physical things do not exist. Copley Square is an example of a transcendent physical thing that does exist.

But you don't buy it do you? Explain why. (I don't buy it either.)

_______________

*Anglosophers use 'abstract'; Eurosophers sometimes use 'ideal.' Same difference (as a redneck student of mine used to say.)

Some Questions about Thinking, Relations, and Relational Expressions

Bill, you said by email earlier that the sentence “Jake is thinking of Zeus” would be true if Jake was indeed thinking of Zeus.

BV: That's what I said, although I would put 'is' where you have 'was.' Is what I said a shocking thing to say?

I have questions for you about the terms ‘obtains’ and ‘satisfies’.

(1) If “Jake is thinking of Zeus” is true, and assuming there is no such thing as Zeus, then does the relation “– is thinking of –” obtain? According to what you said earlier, a relation cannot obtain if any its relata do not exist. But we normally think of a relation obtaining precisely in the case where the sentence which asserts the relation is true. What do you think?

BV: We cannot assume that thinking-of is a relation if every relation is such that its obtaining entails the existence of all its relata. For in the case of Jake and Zeus only one of the relata exists, and it's not Zeus. And yet it is true that Jake is thinking of Zeus. I conclude that the sentence 'Jake is thinking of Zeus,' although grammatically relational, does not express a relational proposition. The sentence needs a truth-preserving analysis that does not commit one to the existence of nonexistent things.

Here are two different candidate analysantia. 'Jake is thinking Zeus-ly.' 'Jake is a Zeus-entertainer.' Neither of these sentences is grammatically relational, and both seem to preserve the truth of the analysandum without commitment to nonexistent things. I do not endorse either analysans.

(2) Is the relational expression “– is thinking of –” satisfied when “Jake is thinking of Zeus” is true? For example, is it satisfied by the two things Jake and Zeus respectively? If not, why not?

BV: No. Why not? Because Zeus does not exist.

Untangling Plato’s Beard

I was asked by a commenter what motivates the thin theory of existence. One motivation is

. . . the old Platonic riddle of nonbeing. Nonbeing must in some sense be, otherwise what is it that there is not? This tangled doctrine might be nicknamed Plato's beard; historically it has proved tough, frequently dulling the edge of Occam's razor. (Willard Van Orman Quine, "On What There Is" in From a Logical Point of View, Harper Torchbook ed., 1963, pp. 1-2)

As I see it, here is how the paradox arises.

1) 'Pegasus does not exist' is true. Therefore:

2) The sentence in question has meaning. (Only meaningful sentences have a truth value.)

3) If a sentence has meaning, then so do its (sentential and sub-sentential) parts. (Compositionality of meaning.) Therefore:

4) 'Pegasus' has meaning. Therefore:

5) Something is such that 'Pegasus' refers to it. ('Pegasus' is a proper name, and the meaning of a proper name is its referent, that to which it refers.) Therefore:

6) 'Pegasus' refers to something that exists. (Everything exists; there are no nonexistent objects; one cannot refer to what does not exist for it is not there to be referred to.) Therefore:

7) Pegasus must exist for it to be true that Pegasus does not exist. Paradox!

None of the first four propositions is plausibly denied. To avoid the conclusion, we must deny either (5) or (6) and the assumptions that generate them. Now Quine is no Meinongian/Wymanian. Quine advocates a Russellian solution which amounts to rejecting (5) by rejecting the assumption that the meaning of a proper name is exhausted by its reference. For Russell, ordinary proper names are definite descriptions in disguise. This allows them to have meaning or sense without reference. Thus 'Pegasus' is elliptical for 'the winged horse of Greek mythology.' This allows the following contextual paraphrase of 'Pegasus does not exist':

It is not the case that there exists an x such x is the winged horse of Greek mythology

which is free of paradox. What the paraphrase says is that the definite description which gives the sense of 'Pegasus' is not satisfied. Equivalently, it says that the concept winged horse of Greek mythology is not instantiated. Thus the original sentence, which appeared to be about something that does not exist but which, if it existed, would be an animal, is really about about a description or concept which does exist and which is assuredly not an animal.

It is a brilliant solution, prima vista. It works for negative general existentials as well. 'Unicorns do not exist,' despite its surface grammar, cannot be about unicorns — after all, there aren't any — it is about the concept unicorn and predicates of it the property of not being instantiated. Extending the analysis to affirmative general existentials, we can say that 'Horses exist,' for example, is not about horses — after all, which horses would it be about? — it is about the concept horse and predicates of it the property of being instantiated.

What about singular affirmative existentials such as 'Harry exists'? Quine maintains that, in a pinch, one can turn a name into a verb and say, with truth, 'Nothing pegasizes' thereby avoiding both Plato's Beard and Meinong's Jungle so as to enjoy, clean-shaven, the desert landscape bathed in lambent light. So what's to stop us from saying 'Something Harry-sizes'? (Quite a bit, actually, but I won't go into that in this post, having beaten it to death in numerous other entries. Briefly, there are no haecceity-concepts: there is no such concept Harry-ness that (i) can exist uninstantiated; (ii) if instantiated is instantiated by Harry and Harry alone in the actual world; (iii) is not instantiated by anything distinct from Harry in any possible world.)

Let us now pause to appreciate what the Russellian (or rather 'Fressellian') approach accomplishes in the eyes of its advocates. It untangles Plato's Beard. It avoids Meinong's jungle. It preserves the existence-nonexistence contrast by situating it at the second level, that of descriptions, concepts, propositional functions, properties, as the contrast between satisfaction-nonsatisfaction (for descriptions), instantiation-noninstantiation (for concepts and properties), and having a value-not having a value for propositional functions, or as Russell puts it, being sometimes true or the opposite.

What's more, it diagnoses the failure of certain versions of the ontological argument. Descartes' Meditation Five version has it that God exists because God has all perfections and existence is a perfection. But if Frege and Russell are right, existence is not even a property of God let alone a perfection of him inasmuch as '. . .exist(s)' has no legitimate use as a first-level predicate and can be be properly deployed only as a second-level predicate. (God is an individual.)

Last, but not least, the Fressellian analysis consigns entire libraries of school metaphysics to he flames, the books in which drone on endlessly about Being and Existence and the distinctio realis, and the analogia entis, and ipsum esse subsistens, ad nauseam. Swept aside are all the hoary and endlessly protracted debates about the relation of essence and existence in individuals: is it a real distinction, and what could that mean? Is it a formal distinction, and what could that mean? Etc. On the Frege-Russell approach there simply is no existence of individuals.

And now you know why the thin theory is called 'thin.' It could also be called 'shallow' in that it eliminates existence as a deep and mysterious topic. The thin theory disposes of existence as a metaphysical topic, reducing it to a merely logical topic. As Quine famously says in an essay other than the one cited above, "Existence is what existential quantification expresses." Thus 'Cats exist' says no more and no less than 'For some x, x is a cat.' You will note that the analysans makes no mention of existence. It features only the word 'cat' and some logical machinery. Existence drops out as a metaphysical topic.

Of course, I don't accept the thin theory; but as you can see, I appreciate what motivates it in the minds of its adherents.

Lukáš Novák on Reference to What is Not

What follows is a re-do of an entry that first saw the light of the blogosphere on the 4th of July, 2014. The draft Lukáš Novák (on my left in the photo) sent me back then for my comments has since appeared in print in Maimonides on God and Duns Scotus on Logic and Metaphysics (Volume 12: Proceedings of the Society for Medieval Logic and Metaphysics) eds. Gyula Klima and Alexander W. Hall, Cambridge Scholars Publishing, 2015, pp. 155-188. I note that our old sparring partner Edward Buckner has an article in this volume, "On the Authenticity of Scotus's Logical Works," pp. 55-84.

………………………………………

Our Czech friend Lukáš Novák sent me a paper in which, drawing upon John Duns Scotus, he rejects the following principle of reference:

(PR) It is impossible to refer to that which is not.

In this entry I will first pull some quotations from Novak's paper and then raise some questions about the view that he seems to be endorsing.

I. Novak's Scotistic View

Novak writes,

Scotus’ position can be simply characterized as a consistent rejection of the PR . . . . According to Scotus, the objects of any intentional relations . . . simply are not required to have any ontological status whatsoever, or, as Scotus puts it, any esse verum. The “being” expressed by the predicates exploited by Francis, like “to be known” (esse cognitum), “to be intelligible” (esse intelligibile), “to be an image of a paradigm” (esse exemplatum), “to be represented” (esse repraesentatum) and the like, is not real or true in any way, irrespectively of whether the relation involved concerns God or man.

[. . .]

It is not necessary to assume any esse essentiae in objects of knowledge: instead, Scotus speaks of “esse deminutum” here, but he points out emphatically that this “diminished being” is being only “secundum quid”, i.e., in an improper, qualified sense – this is the point of Scotus’ famous criticism of Henry of Ghent laid out in the unique question of dist. 36 of the first book of his Ordinatio. If you look for some real being in the object of intellection that it should have precisely in virtue of being such an object, there is none to be found. The only real being to be found here is the real being of the intellection, to which the esse deminutum of the intellected object is reduced.

[. . .]

In other words: if we were to make something like an inventory of reality, we should not list any objects having mere esse deminutum. By speaking about objects in intelligible being we do not take on any ontological commitment (to use the Quinean language) over and above the commitment to the existence of the intellections directed to these objects.

[. . .]

And now the crucial point: it is precisely this intelligibility, imparted to the objects by the divine intellect, what [that] makes human conceiving of the same objects possible, irrespectively of whether they have any real being or not:

[. . .]

In other words: the most fundamental reason why the PR is false is, according to Scotus, the fact that a sufficient condition of the human capacity to refer to something is the intelligibility of that something. This intelligibility, however, is bestowed on things in virtue of their being conceived, prior to creation, by the absolute divine intellect. This divine conceiving, however, neither produces nor presupposes any genuine being in the objects; for it is a universal truth that cognition is an immanent operation, one whose effect remains wholly in its subject (and so does not really affect its object) – in this elementary point divine cognition is not different. Accordingly, objects need not have any being whatsoever in order to be capable of being referred to. (p. 181, emphases added)

II. Some Questions and Comments

As a matter of Moorean fact we do at least seem to refer both in thought and in speech to nonexistent objects and to say things about them, true and false. The celebrated goldner Berg discussed by Bernard Bolzano, Kasimir Twardowski, Alexius von Meinong et al. is a stock example. Suppose that I am thinking about the golden mountain (GM). Since I cannot think without thinking of something, when I am thinking of or about the GM, I am thinking of something. But thinking is not like eating. Necessarily, if I eat something, that thing exists. I cannot eat a nonexistent comestible. Eating takes an existing object; thinking, however, needn't take an existing object. But it must take an object. So it is quite natural to say that in the case before us, the act of thinking is directed to a nonexistent object.

That some objects do not exist (or have any mode of being at all) would seem to following directly from the intentionality or object-directedness of consciousness. My act of thinking about the GM (or about Frodo, to use Novak's example), being intentional is directed to, intends, an object that is not part of the act, but transcendent of it. It follows straightaway that some objects of thinking and linguistic reference have no being. So far, it seems that Dr. Novak is right: we must reject (P) according to which it is impossible to refer to that which is not.

But of course this is puzzling. An object that has no being is nothing. How then can I be thinking about something that is nothing? And if what I am thinking about is nothing at all, then how is my thinking of Frodo different from my thinking of the GM? Acts are individuated by their objects; if the objects are nothing, then they do not differ and cannot serve to individuate the acts trained upon them. What's more, if the GM is nothing at all, then it has no properties; but it does have properties, ergo, etc. So we have an aporetic dyad that needs solving:

a) The GM is something (because every thinking is a thinking of something, and I am thinking of the GM.)

b) The GM is nothing (because there are no mountains of gold in reality outside my mind, nor, for that matter, inside my mind).

If I understand Novak, he wants a theory that satisfies the following desiderata or criteria of adequacy:

D1. Metaphysical possibilism is to be avoided. We cannot maintain that the merely possible has any sort of being. (Novak distinguishes metaphysical possibilism and actualism from semantic possibilism and actualism. Cf. p. 185)

D2. Actualist ersatzism is to be avoided. We cannot maintain that there are actual items such as Plantingian haecceity properties that stand in for mere possibilia.

D3. The phenomenological fact that intentionality is relational and not quasi-relational (etwas Relativliches) as in Brentano is to be respected and somehow accommodated. No adverbial theories!

D4. Eliminativism about intentionality/reference is to be avoided. Intentionality is real!

D5. Nominalist reductionism according to which reference is a merely intralinguistic phenomenon is to be avoided. When I refer to something, whether existent or nonexistent, I am getting outside of language!

Novak does not list these desiderata; I am imputing them to him. He can tell me if my imputation is unjust. In any case, I accept (D1)-(D5): an adequate theory must satisfy these demands. Now how does Novak's theory satisfy them?

Well, he brings God into the picture. Some will immediately cry deus ex machina! But I think Novak can plausibly rebut this charge. If God is brought on the stage in an ad hoc manner to get us out of a philosophical jam, then a deus ex machina objection has bite. But Novak and his master Scotus have independent reasons for positing God. See p. 185. And see my substantial post on deus-ex-machina objections in philosophy, here. Suppose we have already proven, or at least given good reasons for, the existence of God. Then he can be put to work. Or, as my esteemed and fondly remembered teacher J. N. Findlay once said, "God has his uses."

So how does Novak's solution work?

It is sufficient for x to be an object of thought or reference by us that it be intelligible. This intelligibility derives from the divine intellect who, prior to creation, conceives of such items as the golden mountain. But this conceiving does not impart to them any real being. Nor does it presuppose that they have any real being. In themselves, they have no being at all. God's conceiving of nonexistent objects is a wholly immanent operation the effect of which remains wholly within the subject of the operation, namely, the divine mind. The intelligibility is not projected onto items external to the divine intellect. And yet the nonexistent objects acquire intelligibility. It is this intelligibility that makes it possible for us finite minds to think the nonexistent without it being the case that nonexistent objects have any being at all.

This is the theory, assuming I have understood it. And it does seem to satisfy the desiderata (D1)-(D5) with the possible exception of (D3). But here is one concern. We are being told that the intelligibility of the GM, for example, is due to a wholly immanent operation on God's part. That is: no act of divine intellection is directed outward toward a transcendent object even if said object is beingless. But if the divine production of the intelligibility of the GM, say, is wholly immanent then this can only mean that the production proceeds by God's conceiving-GM-ly. But this amounts to adverbialism and a denial of the relationality of intentionality, which Novak is otherwise committed to. Cf. the "pre-philosophical datum" mentioned on p. 186 according to which "we all know that we can refer to non-existing things" such as Frodo Baggins, and yet "we all know that they are not there." Frodo, after all, is purely fictional item "made up" by Tolkien. Talk of reference whether it be thinking reference or reference expressed verbally implies relationality: I am related to what I refer to. But talk of wholly immanent operations of cognition and conception sits none too well with the relational talk of reference.

So my question for Novak is: Did Scotus anticipate the adverbialism of Roderick Chisholm, et al.? Is Scotus an adverbialist?

Here is a second concern of mine. We are told that:

. . . it is a universal truth that cognition is an immanent operation, one whose effect remains wholly in its subject (and so does not really affect its object) – in this elementary point divine cognition is not different. Accordingly, objects need not have any being whatsoever in order to be capable of being referred to.

This implies that both divine and creaturely cognition and conceiving are wholly immanent operations. So what is going on when I think of, or refer to, the GM? It seem that I too would have to be conceiving-GM-ly. But then the objections to adverbialism would kick in.

Here is a third concern not unrelated to the second. The Scotistic-Novakian theory seems to imply that when I think about the golden mountain I am thinking about an operation wholly immanent to the divine intellect. But that is not what I seem to be thinking about. (And how would I gain access to God's mind?) It falls afoul of the phenomenology of intentionality. What I seem to be thinking about has very few properties (being golden, being a mountain) and perhaps their analytic entailments, and no hidden properties such as the property of being identical to an operation wholly immanent to the divine intellect. An intentional object that does not exist has precisely, all and only, the properties it is intended as having.

Connected with this third concern is the suspicion that on Novak's Scotistic theory the act-object distinction is eliminated, a distinction that is otherwise essential to his approach. He wants to deny that merely intentional objects have any being of their own. So he identifies them with divine conceivings. But this falls afoul of a point insisted on by Twardowski.

My merely imagined table does not exist in reality, 'outside' my mind. But it also does not exist 'in' my mind as identical to the act of imagining it or as a proper part of the act of imagining it, or as any sort of mental content, as Twardowski clearly saw. Otherwise, (i) the merely imagined table would have the nature of an experience, which it does not have, and (ii) it would exist in reality, when it doesn't, and (iii) it would have properties that cannot be properties of mental acts or contents such as the property of being spatially extended.

My point could be put like this. The typical merely intentional, hence nonexistent, object such as the golden mountain does not have the nature of an experience or mental act; it is an object of such an act. But if merely intentional objects are divine conceivings, then they have the nature of an experience. Ergo, etc. Novak's theory appears to fall into 'divine psychologism.'

Secure Epistemic Foundations, Language, and Reality

This from Grigory Aleksin:

I have been doing some reading and thinking, and there are a few things that I cannot quite get my head around. I was wondering whether you could help me, or point me in the direction of some work on the issue. My somewhat naive task has been to try and find the most foundational and basic pieces of knowledge that are required by any worldview.

It seems to me there are at least two things that are in some sense foundational:

(1) Something exists

(2) There are correct and incorrect inferences

(1) seems to follow from what is meant by a 'thing' and what is meant by 'exists'. However this is only the case, if there are correct and incorrect inferences. Therefore, (2) is in some sense prior to (1). Hopefully that makes sense.

BV: It does indeed make sense. But I would approach the quest for secure foundations more radically. How do I know (with objective certainty) that something exists? I know this because I know that I exist. 'Something exists' follows immediately from 'I exist.' To say that one proposition follows from another is to say that the inference from the other to the one is correct. The correctness of the inference preserves not only the truth of the premise but also its objective certainty. I agree that your (2) is in some sense prior to (1); it is a presupposition of the inferential move from

(0) I exist

to

(1) Something exists.

My problem arises when I consider that both (1) and (2) are not actually part of reality: both are sentences or linguistic expressions.

BV: Here you have to be careful. Surely a sentence token is a part of reality, even if you restrict reality to the spatio-temporal. The truth that something exists is not the same as its linguistic expression via the visible string, 'Something exists.' That same truth (true proposition, true thought) can also be expressed by a tokening of the German sentence 'Etwas existiert' and in numerous other ways. This suffices to show that the proposition expressed is not the same as the material vehicle of its expression. And already in Plato there is the insight that, while one can see or hear a sentence token, the eyes and the ears are not the organs whereby one grasps the thought expressed by marks on paper or sounds in the air.

So we need to make some distinctions: sentence type, sentence token, proposition/thought (what Frege calls der Gedanke). And this is just for starters.

And should we restrict reality to the spatio-temporal-causal? Are not ideal/abstract objects also real? The sign '7' is not the same as the number 7. A numeral is not a number. I can see the numeral, but not the number. I can see seven cats, but not the (mathematical) set having precisely those cats as members. I can see the inscription '7 is prime' but not the proposition expressed on an occasion of use by a person who produces a token of that linguistic type. The ideal/abstract objects just mentioned arguably belong to reality just as much as cats and rocks.

Thus I have come to consider the role of language. The issue is that language is just a way of mapping reality, and as such is disconnected from it. This raises the question of what 'truth' is, since on one hand we know that there are objective truths, yet truths are only expressed [only by] using language. My question is, then: how can the analysis of language be used to answer philosophical questions? I know that linguistic analysis plays a central role in analytic philosophy, but I cannot help by having [but have] doubts or suspicions that something is wrong. As you see, I cannot fully express what it is that causes me such a headache, but it stems from a suspicion with respect to the use and limits of language, and thus philosophical inquiry.

BV: We do distinguish between WORDS and WORLD, between language and reality. But this facile distinction, reflected upon, sires a number of puzzles. My cat Max is black. So I write, 'Max is black.' The proper name 'Max' maps onto Max. These are obviously distinct: 'Max' is monosyllabic, but no animal is monosyllabic. So far, so good. But what about the predicate 'black'? Does it have a referent in reality in the way that 'Max' has a referent in reality? It is not obvious that it does. And if it does, what is the nature of this referent? If it doesn't, what work does the predicate do? And then there is the little word 'is,' the copula in the sentence. Does it have a referent? Does it map onto something in reality the way 'Max' does? And what might that be? The transcendental unity of apperception? Being? If you say 'nothing,' then what work does the copula do?

One can see from this how questionable is the claim "that language is just a way of mapping reality . . . ." We don't want to say that for each discrete term there is a one-to-one mapping to an extralinguistic item. That would be a mad-dog realism. (What do 'and' and 'or' and 'not' refer to?) Nominalism is also problematic if you hold that only names refer extralinguistically. And you have really gone off the deep end if you hold that all reference is intralinguistic.

Here is another ancient puzzle. A sentence is not a list. 'Max is black' is not a mere list of its terms. There is such a list, but it cannot 'attract a truth-value.' That is a philosopher's way of saying that a list cannot be either true or false. But a sentence in the indicative mood is either true or false. Therefore, a sentence in the indicative mood is not a list. Such a sentence has a peculiar unity that makes it apt to be either true or false. But how are we to understand that unity without igniting Bradley's regress?

And then there is the question of the truth-bearer or truth-vehicle. You write above as if sentences qua linguistic expressions are truth-bearers. But that can't be right. How could physical marks on paper be either true or false?

My question is, then: how can the analysis of language be used to answer philosophical questions?

It is not clear what you are asking. You say that there are objective truths. That's right. Your problem seems to be that you do not see how this comports with the fact that truths are expressed only by using language. The source of your puzzlement may be your false assumption that sentence qua linguistic expressions are the primary vehicles of the truth-values.

Combox open.

A Most Remarkable Prophecy

The Question

Suppose there had been a prophet among the ancient Athenians who prophesied the birth among them of a most remarkable man, a man having the properties we associate with Socrates, including the property of being named 'Socrates.' Suppose this prophet, now exceedingly old, is asked after having followed Socrates' career and having witnessed his execution: Was that the man whom you prophesied?

Does this question make sense? Suppose the prophet had answered, "Yes, that very man, the one who just now drank the hemlock, is the very man whose birth I prophesied long ago before he was born!" Does this answer make sense?

An Assumption

To focus the question, let us assume that there is no pre-existence of the souls of creatures. Let us assume that Socrates, body and soul, came into existence at or near the time of his conception. For our problem is not whether we can name something that already exists, but whether we can name something that does not yet exist.

Thesis

I say that neither the question nor the answer make sense. (Of course they both make semantic sense; my claim is that they make no metaphysical or broadly logical sense.) What the prophet prophesied was the coming of some man or other with the properties that Socrates subsequently came to possess. What he could not have prophesied was the very man that subsequently came to possess the properties in question. This is equivalent to saying that there was no individual Socrates before he came into existence. Before he came into existence there was no merely possible Socrates.

What the prophet prophesied was general, not singular: he prophesied that a certain definite description would come to be satisfied by some man or other. Equivalently, what the prophet prophesied was that a certain conjunctive property would come in the fullness of time to be instantiated, a property among whose conjuncts are such properties as being snubnosed, being married to a shrewish woman, being a master dialectician, being accused of being a corrupter of youth, etc. Even if the prophet had been omniscient and had been operating with a complete description, a description such that only one person in the actual world satisfies it if anything satisfies it, the prophecy would still be general.

Why would the complete description, satisfied uniquely if satisfied at all, still be general? Because of the possibility that some other individual, call him 'Schmocrates,' satisfy the description. For such a complete description, uniquely satisfied if satisfied at all, could not capture the very haecceity and ipseity and identity of a concrete individual.

We can call this view I am espousing anti-haecceitist: the non-qualitative thisness of a concrete individual cannot antedate the individual's existence. Opposing this view is that of the haecceitist who holds that temporally prior to the coming into existence of a concrete individual such as Socrates, the non-qualitative thisness of the individual is already part of the furniture of the universe.

My terminology is perhaps not felicitous. I am not denying that concrete individuals possess haecceity. I grant that haecceity is a factor in an individual's ontological 'assay' or analysis. What I am denying is that the haecceity of an individual can exist apart from the individual whose haecceity it is. From this it follows that the haecceity of an individual cannot exist before the individual exists.

But how could the non-qualitative thisness of a concrete individual be thought to antedate the individual whose thisness it is? We might try transforming the non-qualitative thisness of a concrete individual into an abstract object, a property that exists in every possible world, and thus at every time in those worlds having time.

Consider the putative property, identity-with-Socrates. Call it Socrateity. Suppose our Athenian prophet has the power to 'grasp' (conceive, understand) this non-qualitative property long before it is instantiated. Suppose he can grasp it just as well as he can grasp the conjunctive property mentioned above. Then, in prophesying the coming of Socrates, the prophet would be prophesying the coming of Socrates himself. His prophecy would be singular, or, if you prefer, de re: it would involve Socrates himself.

What do I mean by "involve Socrates himself"? Before Socrates comes to be there is no Socrates. But there is, on the haecceitist view I reject, Socrateity. This property 'deputizes' for Socrates at times and in possible worlds at which our man does not exist. It cannot be instantiated without being instantiated by Socrates. And it cannot be instantiated by anything other than Socrates in the actual world or in any possible world. By conceiving of Socrateity before Socrates comes to be, the Athenian prophet is conceiving of Socrates before he comes to be, Socrates himself, not a mere instance of a conjunctive property or a mere satisfier of a description. Our Athenian prophet is mentally grabbing onto the very haecceity or thisness of Socrates which is unique to him and 'incommunicable' (as a Medieval philosopher might say) to any other in the actual world or in any possible world.

But what do I mean by "a mere instance" or a "mere satisfier"?

Let us say that the conjunctive property of Socrates mentioned above is a qualitative essence of Socrates if it entails every qualitative or pure property of Socrates whether essential, accidental, monadic, or relational. If Socrates has an indiscernible twin, Schmocrates, then both individuals instantiate the same qualitative essence. It follows that, qua instances of this qualitative essence, they are indistinguishable. This implies that, if the prophet thinks of Socrates in terms of his qualitative essence, then his prophetic thought does not reach Socrates himself, but only a mere instance of his qualitative essence.

My claim, then, is that one cannot conceive of an individual that has not yet come into existence. Not even God can do it. For until an individual comes into existence it is not a genuine individual. Before Socrates came into existence, there was no possibility that he, that very man, come into existence. (In general, there are no de re possibilities involving future, not-yet-existent, individuals.) At best there was the possibility that some man or other come into existence possessing the properties that Socrates subsequently came to possess. To conceive of some man or other is to think a general thought: it is not to think a singular thought that somehow reaches an individual in its individuality.

To conceive of a complete description's being satisfied uniquely by some individual or other it not to conceive of a particular individual that satisfies it. If this is right, then one cannot name an individual before it exists.

It Is What It Is

Maybe not. It all depends on what the meaning of 'is' is. (If you are old enough to get the joke, you are old.)

Seriously, though, the above-captioned saying is seeing quite a lot of use lately, or it was ten years ago. It is a sort of present-tensed Que sera, sera. Things are the way they are. Don't kick against the pricks. Acceptance and resignation are the appropriate attitudes.

From a philosophy-of-language point of view, what is interesting is the use of a tautological form of words to express a non-tautological proposition. What the words mean is not what the speaker means in uttering the words. Sentence meaning and speaker's meaning come apart. The speaker does not literally mean that things are what they are — for what the hell else could they be? Not what they are? What the speaker means is that (certain) things can't be changed and so must be accepted with resignation. Your dead-end job for example. 'It is what it is.'

There are many examples of the use of tautological sentences to express non-tautological propositions. 'What will be, will be' is an example, as is 'Beer is beer.' When Ayn Rand proclaimed that Existence exists! she did not mean to assert the tautological proposition that each existing thing exists; she was ineptly employing a tautological sentence to express a non-tautological and not uncontroversial thesis of metaphysical realism according to which what exists exists independently of any mind, finite or infinite.

'What will be will be' is tautologically true and thus necessarily true. What the sentence is typically used to express, however, is the non-tautological, and arguably false, fatalistic proposition that what will be, will necessarily be, that it cannot be otherwise. So not only do sentence meaning and speaker's meaning come apart in this case; a modal fallacy is lurking in the background as well, the ancient fallacy of confusing the necessitas consequentiae with the necessitas consequentis.

Necessitas Consequentiae versus Necessitas Consequentis

More on Tautologies that Ain't: 'He's his Father's Son'

In Vino Veritas

Literally, "in wine, (there is) truth." But the sentence does not bear its meaning on its semantic sleeve. What the familiar Latin saying is used to express, by those who use it correctly, is the thought that a person under the influence of alcohol is less likely to dissemble and more likely to speak his mind and perhaps reveal something that he would not have revealed if sober.

Linguistic meaning, though not reducible to use, cannot be adequately understood apart from use.

Intentionality for Third-World Entities?

Commenter John and I are having a very productive discussion about intentionality. I thank him for helping me clarify my thoughts about this fascinating topic. I begin with some points on which (I think) John and I agree.

a) There is a 'third world' or third realm and it is the realm of abstracta. (I promise: no jokes about Frege's Third Reich. But I can't promise not to speak of Original Sinn and Original Sinn-ers.) Fregean senses, whether propositional or sub-propositional, are abstracta, but not all abstracta are reference-mediating senses. John and I are operating with a provisional tripartite or tri-categorial ontology comprising the mental, the physical, and the abstract.

b) There are instances of intrinsic intentionality. Neither of us is an eliminativist about intentionality in the manner, say, of Alexander Rosenberg. (See Could Intentionality be an Illusion?)

c) There is no intentionality without intrinsic or original intentionality: it cannot be that all intentionality is derivative or a matter of ascription, pace Daniel Dennett. (See Original and Derived Intentionality, Circles, and Regresses.)

d) Nothing physical qua physical is intrinsically intentional, although some physical items are derivatively intentional. (Combine this true proposition with the false proposition that all mental states are physical, and you have an unsound but valid argument for the eliminativist conclusion that there is no intrinsic intentionality.)

Agreement on the foregoing points leaves open the question whether there could be intrinsically intentional abstract items. I tend to think that there are no intrinsically intentional abstract items. John's position, assuming I understand it, is that some abstract items are intrinsically intentional, and that some intrinsically intentional items are not abstract, mental states being examples of the latter.

The bare bones of the debate between John and I may be set forth as an aporetic triad:

1) Fregean senses are intrinsically intentional items.

2) Only conscious items are intrinsically intentional.

3) Fregean senses are not conscious.

It is easy to see that this threesome is not logically consistent: the propositions cannot all be true. John and I assume that the Law of Non-Contradiction holds across the board: we are not dialetheists. So something has to give. Which limb of the triad should we reject? (3) is not in dispute and presumably will be accepted by all: no abstract item is conscious, and senses are abstract. 'Abstract' was defined in earlier entries, and John and I agree on its meaning. The dispute concerns (1) and (2). I reject (1) while accepting the other two propositions; John rejects (2) while accepting the other two propositions.

I argue from the conjunction of (2) and (3) to the negation of (1), while John argues from the conjunction of (1) and (3) to the negation of (2).

My rejection of (1) entails that there are no Fregean senses (Sinne). This is because Fregean senses, by definition, are intrinsically intentional. It follows that they are essentially intrinsically intentional. So if they can't be intrinsically intentional, then they can't exist. Why are senses essentially intrinsically intentional? Well, as platonica, senses are necessarily existent: they exist in all metaphysically possible worlds. It follows that they exist in worlds in which there are no finite minds.* Now a sense, by definition, is a mode of presentation (Darstellungsweise) of its object. It mediates between minds and things. Reference, whether thinking reference or linguistic reference, is routed though sense. The (re)presentational power of a sense is essential to it, and it has this power even in worlds in which there are no words to express the sense, no things to be presented by the sense, and no minds to refer to things via senses. For example, consider a possible world W in which there are no languages, no minds, and no planet Venus. In W the sense that 'Phosphorus' — 'Morning Star,' Morgenstern — expresses in our world exists (because it exists in every world) and has its (re)presentational power there in W. Thus its intentionality is intrinsic to it and does not depend on any relations to words or to things or to minds. It (re)presents non-linguistically and non-mentally and without the need for physical embodiment.

I think it follows that there is no distinction in reality — although there is one notionally — between the power of a Fregean sense to represent and its exercise of this power. There is, in other words, no distinction in reality between the power of a sense to represent and its actually representing. I say this because the existence of what an intrinsically intentional item is of or about has no effect whatsoever on the aboutness of the item. Suppose I am thinking about the Washington monument, but that while I am thinking about it, it ceases to exist. That change in objective reality in no way affects the aboutness of the intentional state. Thus the power of an intrinsically intentional item to represent does not need an external, objectively real, 'trigger' to actualize the power. The extramental existence of the Washington monument is not a necessary condition of the aboutness of my thinking about it. The content and aboutness of my thinking is exactly the same whether or not the monument exists 'outside the mind.' The same goes for senses. The sense of 'Phosphorus' presents Venus whether or not Venus exists. And the content of the sense is exactly the same whether or not Venus exists.

There is an important difference, however, between an intrinsically intentional mental state and a Fregean sense. The occurrent mental state or 'act' — in the terminology of Twardowksi, Husserl, et al. — is the state of a mind. It is the act of a subject, the cogitatio of an ego, where the last three occurrences of 'of' all express the genitivus subjectivus. This is essentially, not accidentally, the case. There has to be an ego behind the cogitatio for the cogitatio to be a cogitatio of a cogitatum. But there needn't be an ego 'behind' the sense for the sense to be a sense of or about a thing. If a Fregean sense mediates a reference between a mind and a thing, it is not essential to the mediation that there be a mind 'behind' the sense.

Here then is an argument against Fregean senses:

4) Every instance of intrinsic intentionality has both a subject and an object.
5) Some instances of reference-mediation by a sense do not have both a subject and an object.
Therefore
6) Some instances of reference-mediation by a sense are not instances of intrinsic intentionality.

When I reject the proposition that Fregean senses are intrinsically intentional items, I thereby reject the very existence of Fregean senses. I am not maintaining that Fregean senses exist but are derivatively intentional items. I do hold, however, that there are derivatively intentional items, maps for example. Maps get their meaning and aboutness from us original Sinn-ers. A map is not about a chunk of terrain just in virtue of the map's physical and geometrical properties. Consider the contour lines on a topographical map. The closer together, the steeper the terrain. But that closer together should mean steeper is a meaning assigned by the community of map-makers and map-users. This meaning is not intrinsic to the map qua physical object. Closer together might have meant anything, e.g., that the likelihood of falling into an abandoned mine shaft is greater.

So some things derive their referential and semantic properties from other things. That is also true of language. Words and phrases don't mean anything in and of themselves. Mind is king: no minds, no meaning. I subscribe to the primacy of the intentional over the linguistic.

John and I agree that Fregean senses, whether propositional or sub-propositional, are explanatory posits. They are not 'datanic' as I like to say. Thus it is a pre-analytic or pre-theoretical datum that the sentences 'The sky is blue' and Der Himmel ist blau 'say the same thing' or can by used to say the same thing. But that this same thing is a Fregean proposition goes beyond the given and enters the explanatory realm. One forsakes phenomenology for dialectics. Now what am I claiming exactly? That there is no need for these posits, that to posit senses is to 'multiply entities beyond necessity in violation of Occam's Razor: Entia non sunt multiplicanda praeter necessitatem? Or am I saying something stronger, namely, that there cannot be any such items as Fregean senses? I believe my view is the latter, and not merely the former. If senses cannot exist, then they cannot be reasonably posited either.

John's view, I take it, is that both Fregean senses and some conscious items are intrinsically intentional or object-directed. He is not maintaining that only third-world entities (abstracta) are intrinsically intentional. By contrast, I maintain that only second-world entities (mental items, both minds and some of their occurrent states) are intrinsically intentional.

I assume that John intends 'intrinsically intentional' to be taken univocally and not analogically. Thus he is not saying that Fregean senses are of or about first-world items in a manner that is analogous to the way second-world items are of or about first-world items.

Fregean senses are intrinsically intentional, necessarily existent, abstract entities. By its very nature a sense presents or represents something apart from itself, something that may or may not exist. It is a natural, not conventional, sign.

Do I have a compelling argument against Fregean senses? Above I mentioned the following argument:

2) Only conscious items are intrinsically intentional.

3) Fregean senses are not conscious. Therefore:

1) It is not the case that Fregean senses are intrinsically intentional.

But this argument appears to beg the question at (2). Why can't there be intrinsically intentional items that are not conscious? If there can be intentionality below the level of conscious mind in the form of dispositionality — see Intentionality, Potentiality, and Dispositionality — why con't there be intentionality above the level of conscious mind in platonica?

Nevertheless, there seems to me something incoherent about Fregean senses. They actually represent even in worlds in which there is nothing to represent and no one to whom to represent. Consider again the sense of 'Phosphorus.' It exists in every world including worlds in which Venus does not exist and no mind exists. In those worlds, the sense in question actually represents but does not represent anything to anyone. It is therefore a non-representing representation, and thus an impossibility.

_____________________

*A finite mind is the kind of mind that needs such intermediary items as Fregean senses or Husserlian noemata to mediate its reference (both thinking reference and linguistic reference) to things that it cannot get completely before its mind in all their parts, properties, and relations. An archetypal intellect such as the divine mind can get at the whole of the thing 'in one blow.' As an infinite mind it has an infinite grasp of the infinitely-propertied thing. An infinite mind has no need of senses. The existence of senses therefore reflects the finitude of our minds. That the reflections of this finitude should be installed in Plato's heavenly place (topos ouranos) seems strange. It looks to be an illict hypostatization. But this thought needs a further post for its adequate deployment.

Excluded Middle, Bivalence, and Disquotation

LEM: For every p, p v ~p.

BV: Every proposition is either true or false.

These principles are obviously not identical. Excluded Middle is syntactic principle, a law of logic, whereas Bivalence is a semantic principle. The first says nothing about truth or falsity. The second does. (See Michael Dummett, Truth and Other Enigmas, Harvard UP, 2nd ed. , 1980, p. xix; Paul Horwich, Truth, Oxford UP, 2nd ed., 1998, p. 79) Though not identical they might nonetheless be logically equivalent. Two propositions are logically equivalent iff each entails the other. Entailment is the necessitation of material implication. Can it be shown that (LEM) and (BV) entail each other? Let's see.

The logical equivalence of the two principles can be demonstrated if we assume the disquotational schema:

DS: p is true iff p.

For example, snow is white is true iff snow is white. Or, if you insist, 'snow is white' is true iff snow is white. In the latter forrmulation, which does not involve reference to propositions, the truth predicate — 'is true' — is merely a device of disquotation or of semantic descent. On either formulation, 'is true' adds no sentential/propositional content: the sentential/propositional content is the same on both sides of the biconditional. The content of my assertion is exactly the same whether I assert that snow is white or I assert that snow is white is true. But if (DS) is granted, then so is:

DS-F: p is false iff ~p.

For example, snow is white is false iff ~(snow is white).

Now if the disquotational schemata exhaust what it is to be true and what it is to be false, then (LEM) and (BV) are logically equivalent.

Given (DS) and (DS-F), we can rewrite (LEM) as

LEM-T: For every p, p is true v p is false.

Now (LEM-T) is simply a restatement of (BV). The principles are therefore logically equivalent given the disquotational schemata.

But this works only if falsehood can be adequately explained in terms of the merely logical operation of negation. This will NOT work if negation can only be explained in terms of falsehood. For then we would enter an explanatory circle of embarrassingly short diameter.

Ask yourself: when is one proposition the negation of another? The negation of p is the proposition that is true iff p is false and false iff p is true. To explain the logico-syntactic notion of negation we have to reach for the semantic notions of truth and falsehood. But then falsehood cannot be exhaustively understood or reduced to negation.

It is telling that to explain negation and the other logical connectives we use TRUTH tables. Such explanation is satisfactory. But it would not be if the redundancy or disappearance or disquotational schemata gave the whole meaning of 'true' and 'false.' (The point is made by M. Dummett, Truth and Other Enigmas, p. 7)

I take this explanatory circle to show that there is more to truth and falsehood than is captured in the above disquotational schemata.

Conclusion: if one's reason for accepting the logical equivalence of (LEM) and (BV) is (DS) then that is a bad reason.

Are there counterexamples to (DS)? It seems to fail right-to-left if 'Sherlock Holmes is a detective' is plugged in for 'p' on the RHS of (DS). Arguably, Holmes is a detective, but it is not true that Holmes is a detective. For it to be true that Holmes is a detective, 'Holmes' would have to refer to something that exists. But this requirement is not satisfied in the case of purely fictional items. I am assuming that veritas sequitur esse, that truth 'follows' or supervenes upon being (existence):

VSE: There are no true predications about what does not exist.

Since Holmes does not exist, 'Holmes is a detective' appears to express a proposition that is neither true nor false. Likewise for its negation, 'Holmes is not a detective.' (LEM) is not violated since either Holmes is a detective or Holmes is not a detective. But (BV) is violated since the two Holmes propositions are neither true nor false.

It is worth noting that from 'Only propositions have truth-values' one cannot validly infer 'All propositions have truth-values.'

Do Fire Alarms Make Assertions?

The Opponent writes,

The alarm means 'there is a fire in the building'. An assertion has taken place, that there is a fire. But it is triggered by a sensor in the building. So asserting is not just something people do.

This is a loose way of talking quite in order in ordinary life, but false if taken literally and strictly. I have no objection to people in ordinary life saying things like, 'The fire alarm is telling us that there's a fire in the building.' But people don't talk like that. You tell me, "There's a fire!" I ask, "How do you know?" You reply, "The fire alarm went off." You DON"T say, "The fire alarm told me so,"or "The fire alarm made an assertion to that effect." You COULD say, "A fireman told me so."

But let's not get hung up in Ordinary Language analysis. The 1950s are long gone.

My claim is that a mechanical contraption cannot make an assertion any more than a 'sensor' can sense anything. Thermostats don't feel heat and smoke detectors do not smell smoke. Oscilloscopes do not detect sine waves; an engineer detects a sine wave by the instrumentality of the oscilloscope. Neither my dipstick nor the oil on my dipstick asserts that there is sufficient oil in the crankcase; I infer that there is from the oil I observe on the dipstick. Inferring, like asserting, is something people do.

All meaning traces back ultimately to Original Meaners, Original Sinn-ers. Am I being too clever for clarity?

A green light means proceed. A red light means stop. But how did those signals come to acquire their conventional meanings? From us, from minds whose intentionality is original, not derived. Surely you don't believe that green, or a green light, intrinsically means that one may proceed.

Let us see if the Opponent and I can find some common ground. I concede that there is a clear sense in which the sounding of a fire alarm means that there is a fire in the building. But this meaning is an instance of derivative, as opposed to original, intentionality. The intentionality derives from us. The sounding of the alarm means what it means only because we have assigned it that meaning. Its intentionality or meaning is thus not intrinsic to it. After all, a fire alarm could be constructed for deaf people that emits a smell instead of a sound, perhaps the awful smell of burnt hair. Obviously, such a smell is not intrinsically significative of anything.

So: if the Opponent concedes that the intentionality of a fire alarm is merely derivative, then we have agreement. If he holds that it is original, then the disagreement continues.

There is a similar pattern with sentences and propositions. I will allow you to say that a sentence is true or false in a secondary or derivative sense so long as you admit that it is propositions that are the primary truth-bearers. Do we have a deal? A declarative sentence is true in virtue of expressing a true proposition.

Atomic Sentences and Syncategorematic Elements

The Ostrich tells me that Frege has no copula. That's not wrong, but there is a nuance that muddies the waters. Suppose Al is fat. The symbolization as Fa suggests the absence of a copula and thus the absence of a syncategorematic element. There appears to be only two categorematic elements, a and F. Well, let's see.

………………………………..

According to Fred Sommers (The Logic of Natural Language, Oxford UP, 1982, 166), ". . . one way of saying what an atomic sentence is is to say that it is the kind of sentence that contains only categorematic expressions." Earlier in the same book, Sommers says this:

In Frege, the distinction between subjects and predicates is not due to any difference of syncategorematic elements since the basic subject-predicate propositions are devoid of such elements. In Frege, the difference between subject and predicate is a primitive difference between two kinds of categorematic expressions. (p. 17)

Examples of categorematic (non-logical) expressions are 'Socrates' and 'mammal.' Examples of syncategorematic (logical) expressions are 'not,' 'every,' and 'and.' As 'syn' suggests, the latter expressions are not semantic stand-alones, but have their meaning only together with categorematic expressions. Sommers puts it this way: "Categorematic expressions apply to things and states of affairs; syncategorematic expressions do not." (164)

At first I found it perfectly obvious that atomic sentences have only categorematic elements, but now I have doubts. Consider the atomic sentence 'Al is fat.' It is symbolized thusly: Fa. 'F' is a predicate expression the reference (Bedeutung) of which is a Fregean concept (Begriff) while 'a' is a subject-expression or name the reference of which is a Fregean object (Gegenstand). Both expressions are categorematic or 'non-logical.' Neither is syncategorematic. And there are supposed to be no syncategorematic elements in the sentence: there is just 'F' and 'a.'

But wait a minute! What about the immediate juxtaposition of 'F' and 'a' in that order? That juxtaposition is not nothing. It conveys something. It conveys that the referent of 'a' falls under the referent of 'F'. It conveys that the object a instantiates the concept F. I suggest that the juxtaposition of the two signs is a syncategorematic element. If this is right, then it is false that atomic sentences lack all syncategorematic elements.

Of course, there is no special sign for the immediate juxtaposition of 'F' and 'a' in 'Fa.' So I grant that there is no syncategorematic element if such an element must have its own separate and isolable sign. But there is no need for a separate sign; the immediate juxtaposition does the trick. The syncategorematic element is precisely the juxtaposition.

Please note that if there were no syncategorematic element in 'Fa' there would not be any sentence at all. A sentence is not a list. The sentence 'Fa' is not the list 'F, a.' A (declarative) sentence expresses a thought (Gedanke) which is its sense (Sinn). And it has a reference (Bedeutung), namely a truth value (Wahrheitswert). No list of words (or of anything else) expresses a thought or has a truth value. So a sentence is not a list of its constituent words. A sentence depends on its constituent words, but it is more than them. It is their unity.

We here touch upon the ancient problem of the unity of the proposition first descried by the immortal Plato.

So I say there must be a syncategorematic element in 'Fa' if it is to be a sentence. There is need of a copulative element to tie together subject and predicate. It follows that, pace Sommers, it is false that atomic sentences are devoid of syntagorematic elements.

Note what I am NOT saying. I am not saying that the copulative element in a sentence must be a separate sign such as 'is.' There is no need for the copulative 'is.' In standard English we say 'The sea is blue' not 'The sea blue.' But in Turkish one can say Deniz mavi and it is correct and intelligible. My point is not that we need the copulative 'is' as a separate sign but that we need a copulative element which, though it does not refer to anything, yet ties together subject and predicate. There must be some feature of the atomic sentence that functions as the copulative element, if not immediate juxtaposition then something else such as a font difference or color difference.

At his point I will be reminded that Frege's concepts (Begriffe) are unsaturated (ungesaettigt). They are 'gappy' or incomplete unlike objects. The incompleteness of concepts is reflected in the incompleteness of predicate expressions. Thus '. . . is fat' has a gap in it, a gap fit to accept a name such as 'Al' which has no gap. We can thus say that for Frege the copula is imported into the predicate. It might be thought that the gappiness of concepts and predicate expressions obviates the need for a copulative element in the sentence and in the corresponding Thought (Gedanke) or proposition.

But this would be a mistake. For even if predicate expressions and concepts are unsaturated, there is still a difference between a list and a sentence. The unsaturatedness of a concept merely means that it combines with an object without the need of a tertium quid. (If there were a third thing, then Bradley's regress would be up and running.) But to express that a concept is in fact instantiated by an object requires more than a listing of a concept-word (Begriffswort) and a name. There is need of a syncategorematical element in the sentence.

So I conclude that if there are any atomic sentences, then they cannot contain only categorematic expressions.

Implication and Presupposition

Dave Bagwill asks:

To be more clear: Do all propositions imply an ontology? Is 'imply' strong enough to bear the weight of 'assertion'? Or is 'imply' basically an equivalent of 'presuppose'?

Still not clear enough. Dave. Not even the third question is clear since you didn't specify the sense of 'imply.' But the third question is clear enough to warrant a brief answer, which is: No. Consider the following which is an intuitively clear example of a proposition resting on a presupposition:

Tom regrets lying to his wife.

Necessarily, if Tom regrets lying to his wife, then Tom has lied to his wife. The antecedent implies (in the sense of 'entails') the consequent. (I have defined 'entails' on many occasions.) But note that it is also true that, necessarily, if Tom does not regret lying to his wife, then Tom has lied to his wife.

This yields a criterion of one type of presupposition. A proposition p presupposes a proposition q just in case both p and its negation ~p entail q. One could also say that an entailment of a proposition p is a presupposition of p if and only if p's presupposition survives the negation of p. (If the preceding sentence does not make sense to you, forget it, and focus on the one preceding it.) Consider now:

Tom is drunk.

Necessarily, if Tom is drunk, then someone is drunk. But it is not the case that, necessarily, if Tom is not drunk, then someone is drunk.

So by the criterion lately enunciated, the 'survival of negation' criterion to give it a name, 'Tom is drunk,' while it implies (entails) that someone is drunk, does not presuppose that someone is drunk.

Therefore, to answer Dave's question, 'imply' (in the sense of 'entails') is not equivalent to 'presuppose.'

Alles klar? Vielleicht nicht!

One could conceivably balk, or baulk in the case of the Bad Ostrich, as follows: It is not clear, or it is false, that if Tom does not regret lying to his wife, then Tom has lied to his wife. The Ostrich could say, "Tom does not regret lying because he didn't lie in the first place."

As you can see, the topic of presupposition is a murky one, and part of the murkiness is due to the fact that presupposition is at the interface of the semantic and pragmatic, and it is not clear how they gear into each other, if you will excuse the mixed metaphors.

More on Assertion and Presupposition

I continue to worry this technical bone, which is not a mere technicality, inasmuch as the topic of presupposition opens out upon some very Big Questions indeed. Anyway, back to work. I thank Ed Buckner for getting me going on this.

…………………

It should be obvious that one does not assert everything that the content of one's assertion entails. If I assert that Venus is a planet, I do not thereby assert that either Venus is a planet or Putin is a former KGB agent, even though the content of my assertion entails the disjunctive proposition. The content of an assertion is a proposition, and for any proposition p, p entails p v q.

A more interesting, and more difficult, question is whether one asserts any proposition that the content of one's assertion entails (apart from the proposition that is the content of the assertion).

Suppose you ask who won the 10K Turkey Trot and I assert that Tony won the race. Do I thereby also assert that he competed in it? That he competed in it is entailed by the fact that he won. And it is entailed in a stronger sense that the sense in which Venus is a planet entails Venus is a planet or Putin is a former KGB agent. For there is a semantic connection between winning and competing, but no semantic connection in the Venus-Putin case. You could say that it is analytically impossible that Tony win without competing: what makes it true that there is no possible world in which Tony wins but does not compete is the semantic connection between winning and competing.

Still, I want to say that Tony's competing is presupposed but not asserted when I assert that he won the race. Necessarily, anything red is colored. But when I assert that Tom the tomato is red, I do not thereby assert that it is colored, although of course I presuppose that it is colored. Note the word 'thereby.' It is no doubt possible for me to assert that Tom is colored, a 'vegetable of color' if you will, but that is a different assertion.

Go back to Tony the runner. That Tony did not cheat by taking a short cut is analytically entailed by the fact that he won. (To win a foot race it does not suffice to be the first to cross the finish line. Remember Rosie Ruiz of Boston Marathon 1980 notoriety?) Will you say that when I assert that Tony won the race I also thereby assert that he did not cheat by taking a shortcut? I would say No. For that would be an unbearably counter-intuitive thing to say. I presuppose, but do not assert, that Tony did not cheat by taking a shortcut

You can see how this series of questions can be extended. One can cheat by getting a head start or by jumping in at mid-course, which is what Rosie Ruiz did at Boston. You can cheat by hiring a a world-class doppelgaenger, by wearing special shoes . . . .

Note also that if Tony won, it follows that he either won or didn't win. Will you say that when I assert that Tony won the race I am also thereby asserting that he either won it or didn't? When I assert that Tony won, I am not asserting the Law of Excluded Middle (LEM). At most, LEM is a presupposition of my assertion, and of every assertion.

If Tony won, then it was possible that he win. For everything actual is possible. But when I assert that Tony won, I presuppose, but do not assert, that it was possible at the time of the race that Tony win.

I am toying with a strong thesis:

When an agent A makes an assertion by uttering or otherwise tokening a sentence s (which is typically, but needn't be, in the indicative mood), the content of the assertion is exactly the (Fregean) proposition explicitly expressed by the tokening of s and no other proposition. Propositions other than the content proposition that are entailed by the content proposition are at most presuppositions of the assertion.

Why hold this view? Well, it seems to me that what I assert on any occasion is precisely what I intend to assert on that occasion and nothing else. When I make an assertion I translate into overt speech a belief that I have. The content/accusative of the belief is a Fregean proposition and there is nothing in that proposition that is not open to my mind at the time I express my belief.