Hence, according to Devitt, "no doctrine of truth is constitutive of realism". — Banno

Which is what I am badly saying elsewhere.

So to the first section, in which Devitt characterises realism as the view that physical entities exist independently of the mental. Devitt notes with considerable glee that there is nothing in this definition about truth. He goes on to point out that truth is independent of the evidence at hand. "Truth is one thing, our means of discovering it, another". Hence, according to Devitt, "no doctrine of truth is constitutive of realism". — Banno

I read the SEP article, and if I was to align myself with anyone it would be Devitt. Truth it seems is a property of language (proposition claims). However, reality (that which exists independent of the mind) is not about what we claim (what we claim may or may not align with reality), but about facts or states-of-affairs (things that are generally considered to be independent of minds) quite apart from what goes on in language. This has been my understanding of this argument, and I believe it's why Devitt says, "no doctrine of truth is constitutive of realism."

Doubtless, but I have been unable to follow your argument, especially in your new thread.

Tying the discussion to an article may help.

if I was to align myself with anyone it would be Devitt. — Sam26

yes, but curiously there is this from the first page of the article:

Yet I suspect that many philosophers are skeptical about Dummett's argument: it smacks too much of positivism and Wittgensteinianism

Something is odd here.

Yet I suspect that many philosophers are skeptical about Dummett's argument: it smacks too much of positivism and Wittgensteinianism

Something is odd here. — Banno

Yes, I noticed that too. Devitt seems to mischaracterize Wittgenstein, at least it seems that way from this statement. Positivism doesn't equate to Wittgensteinianism, if that's his point. Although I'm not completely sure of his point.

I'm now reading the paper by Devitt.

Dummett's arguments are confusing, to say the least. If you can't be clear, one wonders if you have a good argument.

Yet I suspect that many philosophers are skeptical about Dummett's argument: it smacks too much of positivism and Wittgensteinianism

Something is odd here. — Banno

That's more than odd and doesn't inspire confidence.

What an excellent paper. Frankly Devitt could have stopped the paper at:

"Realism says nothing about truth nor even about the bearers of truth, sentences and beliefs (except perhaps, in its use of 'objective', the negative point that beliefs do not determine existence). Realism says nothing semantic at all" - and be done with it.

It has always struck me as odd that realism ever turned upon some human activity like truth-telling at all. It has always been the status of truth (what kind of thing is truth) and not a 'theory of truth' which any 'realism' would need to tackle.

Devitt's also right to point out that Dummett's appropriation of Witty's 'meaning is use' is somewhat underhanded. To take that phrase seriously would be precisely to rule out Dummett's project. That Dummett has been taken so seriously at all should be puzzling, were it not for the prevalence of 'realist theories of truth' before him.

The causal talk seems a bit iffy to me, but that's not the point of the paper I guess.

:up:

Section I.3 is mislabeled III.3 in the paper.

In that section Devitt claims that Dummet's approach is a direct transfer form his metaphysics of mathematics.

Summarising the arguments from I.3.

i. The disagreement between platonists and intuitionists in maths concerns the truth conditions for mathematical statements
ii. There is no further argument, in particular about the substantive nature of mathematical statements - the difference is metaphorical.
iii. This view can be extended to other fields.

Together these provide the reason for thesis (A), that realism comes down to a choice between verification and "evidence - transcendent" truth.

Devitt argues that each of these is problematic, and further that even if they were granted, there remains an unclosed gap between truth being understood thus and objects as being mind-independent.

Doesn't Davidson use truth conditions as meaning? If you jettison that, can you still use Davidson to avoid propositions?

And if not, are you prepared to accept platonist abstract objects (and those weird unstated statements)?

To quote:

Realism (as I have defined it), requires the objective independent existence of common-sense physical entities.

- Devitt

I saw the same trick with Witty 'defining' language as X then restating the same definition (under words) to dispute the existence of a Private Language.

Of course, if you define something from the get go as X it cannot be anything other than X.

Regardless of whether or not the metaphysical realist explicitly claims to be a semantic realist when he argues for the “objective independent existence of common-sense physical entities,” it may be, as Dummett believes, that this position entails semantic realism.

If these independent entities are the things we talk about and the things that determine our statements to be either true or false then is this not recognition-transcendent and bivalent truth-conditions?

Or would you say that metaphysical realism as Devitt describes it is compatible with semantic antirealism as Dummett describes it? Is the “objective independent existence of common-sense physical entities” compatible with something like a verificationist account of meaning and truth? If not then proving the latter disproves the former.

Banno seemed to have recognised how the issues are tied in the previous topic. He started by saying:

Speaking very roughly, just to get started, realism holds that ...stuff... is independent of what we say about it; anti-realism, that it isn't. — Banno

And then continued with:

Stealing blatantly from my Rutledge Encyclopaedia of Philosophy, a realist may hold to things like that correspondence to the facts is what makes a statement true; that there may be truths we do not recognise as such, do not believe and do not know; that the Law of excluded middle holds for things in the world; and that the meaning of a sentence may be found by specifying it's truth-conditions. An ant-realist may in contrast hold that truth is to be understood in sophisticated epistemic terms, perhaps as what a "well-conducted investigation" might lead us to believe; that there can be no unknown truths; that we need include "unknown" as well as true and false in our logical systems; and that the meaning of a sentence is to be found in what it might assert. — Banno

Or would you say that metaphysical realism as Devitt describes it is compatible with semantic antirealism as Dummett describes it? Is the “objective independent existence of common-sense physical entities” compatible with something like a verificationist account of meaning and truth? If not then proving the latter disproves the former. — Michael

I think we could just deflate talk of unknown truths. People who speak of 'searching for the truth' aren't committing to unspoken propositions hanging out there in limbo.

They're just saying they're looking for evidence of something. They're actually verificationists.

II Conjectures B and C

The flow of the argument is convolute. The section appears to flow as follows:

• strangeness of equating a semantic with a psychological dispute (p.83)
• division of B and C into B1, C1 and B2, C2. (p.84)
• reliance on propositional assumption in B1 & C1 (p.84-5)
• lack of evidence for the propositional assumption (p.86-7)
• Argument agains the propositional assumption (p.88-90)

By the end of that section Devitt has rejected the line of argument in B1 & C1.

Note that Devitt advocates the use of truth conditions as explaining meaning, in terms of reference in an apparently extensional interpretation, together with a causal approach to reference (top p.89), but rejects thinking of the meaning of a sentence as an entity (bottom p. 86).


There's more going one here than just either/or.

I've only skim read the above article, so correct me of i'm wrong, but i'm of the impression that Dummett (according to Devitt) is equating, or maybe even equivocating, intuitionism with intuitionistic logic. Under the latter interpretation it is indeed the case that the truth of a proposition is synonymous with, and describable in terms of, the sole activities of the mathematician. One can say in this case that the syntax is the semantics. But under the former interpretation the issue is more subtle and already accommodates my (potentially incorrect and off-base) understanding of Devitts position.

According to intuitionistic logic and intuitionistic type theory, the construction of anything, including the natural numbers and arithmetic, is exhaustively described by axioms and rules of inference, in that terms in the logic make no reference to contingent events of the outside world. So for example, the definition of "one" is reducible to an infallible function called "Successor" operating on a directly observable state "O" to yield in every case a directly observable state "SO". ( The infallibility here looks suspiciously platonistic if the logic is interpreted as being literally true, as opposed to interpreted as being an approximate model of something or a set of normative principles).

But according to intuitionism objects and numbers can also be lawless, where an object is said to be "lawless" if it's existence and/or value isn't decided by the formal system it is part of, but by something not described by, and external to, the formal system.

For example, the standard model of Heyting-arithmetic, which is the intuitionistic logic equivalent of Peano arithmetic, does not include terms for representing the random outcome from physically tossing a die, whereas a corresponding system in intuitionism can, where such a term is said to be "lawless", meaning that the term refers to a value that isn't internally decided by the logic and whose value only potentially exists. In software engineering, such objects are often called "Promises" and are used by programs to denote external random events of the future, such as unreliable and uncertain server responses, where the possibilities of successful replies, failed replies and no replies must be handled by the program logic.

It should also be remarked, that the founder of intuitionism, Brouwer, considered mentally-created numbers as being lawless in so far as they aren't consciously associated with the outcome of a formula.
So by this philosophy, a "lawful" number is merely a value that is logically interpreted, either consciously or practically, as being of the codomain of a formula.

In summary, it isn't the case that intuitionism equates truth with operational construction as in intuitionistic logic. Intuitionism also generalises the intuitionistic-logic notion of proof to include terms that are merely potentially referring, and that when referring refer to contingent events external to the prior observations, construction practices and prior knowledge of the creating subject. But at the same time, intuitionism does not assume that such potentially referring terms are actually referring until as and when the terms are externally initialised with values. Therefore according to Quine's maxim "to be is to be the value of a bound variable", intuitionism cannot be described as realistic.

I get the initial impression that Devitt is arguing for a very weak form of "realism" that merely denies the reduction of truth to principles of construction stateable by a priori laws. Nevertheless, to my understanding his views seem to be already accommodated by intuitionism, which isn't considered to be a realist philosophy of mathematics.

My questions are therefore:

Every computer program that interacts with the world is describable in the language of intuitionism, in which terms presently either have values or don't have values. What in addition, if anything, does Devitt think needs including?

Is Devitt's very weak realism, as i understand it, really acceptable to common-sense realists? how is it different to negative theology?

But according to intuitionism objects and numbers can also be lawless, where an object is said to be "lawless" if it's existence and/or value isn't decided by the formal system it is part of, but by something not described by, and external to, the formal system. — sime

Lots of great things about your post. I did a really bad job of trying to highlight truth of a system and truth external to the system in another post and how when those two systems share a one way relation (external can set values internally but internal cannot set values externally), one can construct a variety of internal systems to predict/interpret the inputs from the external system, and choose the system that works best in the context employed rather than attempt to have the internal system account wholly for the lawless external values.

To an extent that might be right, noting that Devitt is criticising Dummett for moving from the metaphysics of mathematics to the metaphysics of the world without supporting justification.

I'm not sure why you characterise Devitt's realism as "weak".

III Competence as a practical ability

Considering now B2 and C2

• Dummett must show that all truth conditions are verificationist (p.90)
• Devitt's version of Dummett's argument as parsed without using knowedge; (p.91)
• Dummett rejects holism p.(92)
• Dummett relies on a positivist epistemology (p.93)
• Dummett relies on a description theory of reference (p.94)
• Dummett's claims to "meaning is use" are misplaced (p.95)
• Dummett's theory of learning language is too passive. (p.96
• There may be conditions that are associated with a sentence and yet which go unrecognised by the speaker (the truth conditions might be incomplete and yet successful?) (p.96)

In regard to the argument around p. 95.

Devitt sees Dummett as adopting the slogan "meaning is use"; I've recollections of this being dismissed as a mere slogan from a few years after this paper was published; perhaps this is the origin of that criticism.

There's this paragraph:

Wittgenstein'slogan has to be construed in a Wittgensteinian way to give the required support: 'use' must be taken to mean "recognizable conditions of conclusively justified use." That is how Dummett does construeit. The step from the indubitable fact to this construal is a giant step. What justifies it?

That is, Devitt and Dummett seem to both see Wittgenstein as advocating that the use of a sentence is the recognisable behaviour with which it is associated - the behaviourist criticism.

Devitt goes on to explain that treating use as behaviour is insufficient. I agree, but further, I think Wittgenstein might also have agreed. Wittgenstein is at pains to point out the open-ended nature of language, the way in which it adapts and changes to the needs of those using it; but also the capacity of language to indicate, if not actually say, what is outside of mere stating.

Hence the rather rude accusation that Dummett's argument "smacks too much of positivism and Wittgensteinianism" is based on too-small a reading of Wittgenstein.

Added:

Devitt's also right to point out that Dummett's appropriation of Witty's 'meaning is use' is somewhat underhanded. — StreetlightX

Indeed, I see what you meant now.

IV. Verificationisim and Realism

• realism, truth and understanding are distinct; they can be related by inference, not entailment (p.97)
• Realism has primacy (p.97-8)
• The theory of understanding already assumes a world to be understood, and so is the wrong place to start. (p.98)
• Even if an inference from understanding to antirealism were granted, the link between realism and truth would remain. (p.98)
• A verificationist theory of understanding yields realism (Lees-Quine) as well as antirealism

What an excellent paper. — StreetlightX

Do you have any significant problems with the following , or are you in general agreement with its sentiments?

‘[tokens] of the most commonsense, and scientific, physical types objectively exist independently
of the mental. Realism about ordinary objects is confirmed day by day in our experience . . . Given this strong case for Realism, we should give it up only in the face of powerful arguments against it and for an alternative. There are no such arguments.’

of the most commonsense, and scientific, physical types objectively exist independently
of the mental. Realism about ordinary objects is confirmed day by day in our experience . . . Given this strong case for Realism, we should give it up only in the face of powerful arguments against it and for an alternative. There are no such arguments.’ — Joshs

It doesn't work that way. There is no "case" for realism in everyday experience. Realism is a purely philosophical thesis outside that orbit.

So no, the opposition to realism carries no burden, and since idealists have been around for thousands of years, it can't be argued that there's anything extraordinary about it.

Both realism and anti-realism are dogmas.

Seems alright to me.

‘[tokens] of the most commonsense, and scientific, physical types objectively exist independently of the mental. Realism about ordinary objects is confirmed day by day in our experience . . . Given this strong case for Realism, we should give it up only in the face of powerful arguments against it and for an alternative. There are no such arguments.’ — Joshs

What strikes me as ironic is this would be the view my uneducated grandmother would hold. It's very commonsensy and relies on a very literal interpretation of experience.

What strikes me as ironic is this would be the view my uneducated grandmother would hold. It's very commonsensy and relies on a very literal interpretation of experience. — Tom Storm

What a coincidence. I was just reading a review of a compilation of essays by Devitt, which concluded:

‘Putting Metaphysics First is a nicely written defence of (what Jerry Fodor might call) Granny’s philosophy.’

What strikes me as ironic is this would be the view my uneducated grandmother would hold. — Tom Storm

Realism has to be un-learned. It takes sophistication. Idealism, more so.

To assert realism implies an understanding of anti-realism.

Doesn't Davidson use truth conditions as meaning? — frank

Sort of. If you know the truth conditions for a statement, what more could you need?

Well, quite a bit, as his holism indicates.

If you jettison that, can you still use Davidson to avoid propositions

Who is avoiding propositions?

Who is avoiding propositions? — Banno

Davidson did. I would think a realist would want to avoid them. They're 'meaning as an entity.' Abstract objects.

They're 'meaning as an entity.' Abstract objects. — frank

Are they? I would have said they were just statements that are either true or false. I had no idea that they were magical.

So the proposition is what is common between "Schnee ist weiss" and "Snow is white"? But isn't that just that both are true only if snow is white? Arn't hey both used in much the same sort of way, but not quite?

Why do we need propositions?