The first few paragraphs, then.

Davidson begins by characterising the notion of conceptual scheme he wishes to critique. A conceptual scheme is such that what counts as real is relative to the scheme, because the scheme supposedly organises and categorises our experiences. Hence, what is said in one scheme is incommensurable with what is said in some other scheme, since any standard that might be used to relate one scheme to another is itself part of one scheme or another.

Notice that he is not arguing that this is the case, but setting out the characteristics of the notion of conceptual scheme to which the article is being addressed.

The obvious question is, are there advocates of such schemes? Whorf and Kuhn are mentioned later. Feyerabend, arguably. Epistemic relativism is not at all uncommon on this forum.

Davidson gives examples of impressively different descriptions, but notes that these differences are set out in a single language.

This presages the strategy he is going to employ; that the claimed incommensurable descriptions are set out in the one language - in this case, English; and that hence, they are not incommensurate.

SO Davidson sets up a paradox. On one hand is the notion that there are points of view so radically different as to be incomparable. On the other, we could only be aware of this by comparing both points of view.

He then proposes that translation provides a way to compare conceptual schema. Not translation merely from one language to another, since folk with differing languages may share a conceptual scheme; but translation from one conceptual scheme to another. In this way one might be able to identify each conceptual scheme in terms of the way it translates, one into the other.

There's a possible objection here in that one might argue that conceptual schemes somehow inhere in the mind without or before language. Sometimes @creativesoul seems to think something like this. I'd suggest that if this were so, then either this purely mental stuff can be translated into our everyday language, in which case its purpose is lost; or if the mental stuff cannot be translated into our everyday language, then it are irrelevant to the discussion, dropping out like a boxed beetle.

Davidson takes a paragraph to set aside mystical considerations. He seems to find them incoherent.

So we find, if we accept these considerations, that language cannot be set aside unless one also sets aside ones conceptual scheme; we find ourselves in the position of equating translation of language with translation of conceptual scheme. We may say that if what one person holds to be the case cannot be translated for another person, then these two folk hold to distinct conceptual schemes.

We have then a way of elucidating, and differentiating, conceptual schemes based on translation.

Can translation fail totally?

There's then a glorious compression of Davidson's conclusion into a single sentence, containing no less than four negations.

Suppose one sees a certain activity, and wants to know if it is language behaviour or not. Could one recognise a behaviour as language use without also understanding what was being said? Could a cetologist conclude that some given dolphin behaviour was language without also being able to say what the dolphins were discussion - at least in the most general terms?

It seems not. But Davidson is not happy just to accept this, he wants to present an argument for it.

Could one recognise a behaviour as language use without also understanding what was being said? — Banno

I know when a dolphin is trying to talk but I don't always know what a dolphin is saying. Also a young child. Also (I do life enrichment in long-term care) stroked-out aphasics. But go on.

And that's why Davidson does not use that argument, but only mentions it here as a conclusion.

Yes. A brief minor tangent. No problem. I'm more interested in the meat of the scheme-content dogma.

I read the stuff about scheme-content dogma as another side-issue, part of the set up for the discussion of the implications of T-sentences.

That makes sense, considering your fondness for them. How do you read it?

Davidson goes off in search of a way to analyse translation. He's heading to T-sentences, but he takes his time.

There's a brief mention that being able to understand a language seems very close to being able to attribute complex attitudes to folk.

He considers the transitivity of language; language A translates into language B, and so on, until at language X there is no longer anything left from language A. HE raises the issue of how we would know that someone along this path was engaged in translation... how would we be able to tell that Fred was translating language W into language X unless we already understood one or the other? But he rejects this path, too.

Davidson then makes an interesting distinction between two sorts of possible worlds...

In one, we alter what is the case in this world in order to construct other possible worlds. This I take to be the usual brief of modal logic. "...using a fixed system of concepts... we describe alternate universes"; and is associated with a dualism of necessary and possible sentences.

In the other, which he attributes to Kuhn but for which we might blame quite a few others, what is the case is held steady while those observing create their own conceptual worlds. This forces a divide between scheme and content.

All sorts of traps here, too; the difference between analytic-synthetic and necessary-possible, for one.

After a quote from Feyerabend he considers what would be involved in the case where the content was held firm while the conceptual scheme changes, and one finds oneself in a different world.

A favourite argument of mine comes next, one I have borrowed many times, so I will quote at length:

Suppose that in my office of Min- ister of Scientific Language I want the new man to stop using words that refer, say, to emotions, feelings, thoughts and inten- tions, and to talk instead of the physiological states and happen- ings that are assumed to be more or less identical with the mental riff and raff. How do I tell whether my advice has been heeded if the new man speaks a new language? For all I know, the shiny new phrases, though stolen from the old language in which they refer to physiological stirrings, may in his mouth play the role of the messy old mental concepts.

Thus falls the Churchland's attempt to eliminate folk psychology.

Davidson appears to take analytic-synthetic and scheme-content to be much the same distinction here. I don't quite follow that.

But I will happily join him in rejecting the analytic-synthetic distinction. All language is post-hoc, if you will; none of it has meaning without context; none is known a priori.

So Davidson comes to consider a definition of language and conceptual schemes that is not reliant on translation; that there is stuff, the stuff needs sorting, and the conceptual scheme is what does that sorting. Davidson is here squashing a vast range of philosophical ideas into few small paragraphs; and in summarising it I am insulting it further. Experience, sense data, phenomena, prediction, the given...

He sorts the sorting into two sorts... organising and fitting. Organising is dividing the stuff up into simples and complexes. Fitting involves more interaction, or at least further iteration, as stuff is sorted to suit one's needs, attitudes, or whatever.

He then first considers organising stuff, and next, fitting stuff.

Doubtless in making Davidson's path clear to myself here, I've lost what very few readers remained.

On the edge of my seat here.

So we have a distinction between stuff, and language as sorting that stuff by organising it. Conceptual schema would then be the sorting.

So incommensurable schema would be incommensurable sorting of the same stuff.

Now one might agree that we could disagree as to how to sort this or that; but it makes no sense to suggest we disagree about everything. Davidson talks of organising the closet as opposed to organising the shirts in the closet - it's much the same point as Wittgenstein made in On Certainty; one might doubt something, but one cannot sensibly doubt everything; since then one must doubt the very stuff that makes doubt possible.

So one cannot make sense of incompatible schema in this way.

Davidson appears to take analytic-synthetic and scheme-content to be much the same distinction here. I don't quite follow that. — Banno

This paper may help to clarify that bit. Are you familiar with it?

"Two Dogmas of Empiricism"

http://www.ditext.com/quine/quine.html

Then we have the stuff sorted by how we interact with it - how it fits.

Davidson argues that to fit is to be true. Pretty simple.

And this is the bit that @ZzzoneiroCosm questioned earlier in this thread.

I see the argument here as a rejection of defining truth in terms of fit. Davidson takes truth as fundamental. That's why fitting the facts adds nothing to being true, why fitting the conceptual scheme is likewise pointless.

And I agree with this. Theories of truth are fraught, because we already know what it is to be true in any given case, and hence any further discussion can only detract from our understanding.

Or, if you prefer... and so to T-sentences.

Yep. Davidson was Quine's student, and is considered to be continuing Quine's program.

Dang, dropped my slide. that's the trouble with glass. Not a good day - seem to have broken the ground wire in my Gretsch as well.

I'm off to do other things for a while. Questions and corrections welcome. Also, what bits of what I have written are least clear?

I always preferred glass though. Just need a back-up.

It's all clear to me. I'll probably won't say much till after the T-sentence bit. I did note this inconsistency in Davidson, which may have some connection to your take on the T-sentence:


Davidson's inconsistency relieves the pressure of the most troublesome paragraph, quoted above:

First Davidson says:

"Nothing, however, no thing, makes sentences and theories true..."

But he closes by saying:

"...the familiar objects whose antics make our sentences and opinions true..."

Is it nothing or is it objects (specifically, their "antics") that make our sentences true? This looks like a significant inconsistency.

Our attempt to characterize languages or conceptual schemes in terms of the notion of fitting some entity has come down, then, to the simple thought that something is an acceptable conceptual scheme or theory if it is true

T-sentences are central to the way Davidson treats truth, and indeed meaning.

And the criterion of a conceptual scheme different from our own now be- comes: largely true but not translatable.

So an incommensurable conceptual schema will have to be both true and not translatable.

You are talking about models of perception, yes? The notion that an organism builds an internal image of what is around it, in order to better choose pathways and so on? — Banno

Not entirely, you'd need to read the Friston paper to get a full idea, but I understand it's not everyone's cup of tea, so I'll try to summarise. It's not that an organism builds and internal image of what's around it, so much as the organism has mathematical models of the likelihoods of various causes of the sensations which derive from what's around it. When I say model, I mean mathematical, or at least computational. Davidson seems to see conceptual schema (or at least the claim of conceptual schema which he is dismantling) as a filing cabinet with all the content (the way the world is) filed away. I see conceptual schemes more as rules for behaviour, not content-mediated at all. I don't think our brains are like libraries at all, they're inference machines, actively trying to minimise variance between expectation and sensation, so when Davidson concentrates of the 'storage' aspect (sorting and fitting) he's missing much of where computational neuroscience has gone nowadays

Mystical, hidden stuff... how do we talk about that? — Banno

Inference.

___

Really nice summary of the paper so far in your latest posts, thanks. If only every thread could go like that - interesting paper > clear exegesis > discussion - we'd have a marvellous forum.

So we have a distinction between stuff, and language as sorting that stuff by organising it. Conceptual schema would then be the sorting. — Banno

I think this is about right, but I think it misses an elegant bit about the argument. We don't need to know precisely how conceptual schemes work so long as they place necessary constraints on language use. If you can look for the necessary indicators that would be there in language use if conceptual schemes differed wildly, then you don't need to talk about conceptual schemes in general, just about how they should impact language use if they're there at all.

So "language use including interpretation thereof not behaving like there's a scheme-content distinction" is the proposed defeater of "there are wildly varying conceptual schemes that lead to untranslateable sentences between agents that use/have those schemes".

The discussion could be more broad, but it need not be. It's like... propositions and sentences as a minimal worked example.

But I will happily join him in rejecting the analytic-synthetic distinction. All language is post-hoc, if you will; none of it has meaning without context; none is known a priori. — Banno

This caught my eye. Pointing with words, commanding, asking, apologizing, condemning, etc. aren't post hoc. They're actions as much as running or climbing.

Analysis is post hoc. Identifying the use of logic is post hoc. Separating words from meaning (which is required for translation) is post hoc.

Per Quine, the ability to apply logic to new situations has to be innate. It can't be learned.

Right?

Thanks for the exposition @Banno.

Being somewhat concrete and visual in my thinking, I like to see how things work in practice. So here's one I watched earlier.

There are two aspects that relate somewhat, that I want to mention, and anyway it's a fascinating program if you have access.

The first is a matter of translation. The Himba are said to have 'marriages' but they differ from the Christian tradition in being polygamous, and more of a social organisation than anything remotely romantic or even sexual. Other relationships are translated as having 'girlfriends' and 'boyfriends', and one woman talked quite openly and matter-of-factly about her boyfriend being the father of her most of her children. Marriage for the Himba man seems to be mainly a matter of having someone-or two or three to cook for him. It's a social relation for which there is no English word, and we use 'marriage' with a new meaning that becomes gradually clearer as one watches the programs.

So we find, if we accept these considerations, that language cannot be set aside unless one also sets aside ones conceptual scheme; we find ourselves in the position of equating translation of language with translation of conceptual scheme. We may say that if what one person holds to be the case cannot be translated for another person, then these two folk hold to distinct conceptual schemes. — Banno

So all I want to say is that in practice, translation has to involve a learning of a new culture - my brief characterisation here is as inadequate as the translation it points out the inadequacy of.

The second aspect has to do with those 'deep biases' of the article linked above, and relates to certain threads current here.

The show has attracted some negative criticism. The representation of Himba participants has been described as racist and exploitative, while the Moffatt family were seen as selfish and ignorant.

Yet while critics of the show may be well meaning, their views portray the Himba as passive victims with neither agency nor power, and the Moffatts as uneducated intruders. This reveals a deeply embedded paternalistic and imperialistic view of both Indigenous and working-class people and highlights why we need TV shows that challenge these biases.

My view of the article's view of the critic's view of the program's portrayal and treatment of a culture, or rather of two cultures, is that there is indeed 'a measurement problem' that is very real, even if the very idea of it is incoherent.

So "language use including interpretation thereof not behaving like there's a scheme-content distinction" is the proposed defeater of "there are wildly varying conceptual schemes that lead to untranslateable sentences between agents that use/have those schemes". — fdrake

Nothing is in principle untranslatable, but in practice in so many cases, life's too short.

He then proposes that translation provides a way to compare conceptual schema. Not translation merely from one language to another, since folk with differing languages may share a conceptual scheme; but translation from one conceptual scheme to another. In this way one might be able to identify each conceptual scheme in terms of the way it translates, one into the other.

There's a possible objection here in that one might argue that conceptual schemes somehow inhere in the mind without or before language. Sometimes creativesoul seems to think something like this. I'd suggest that if this were so, then either this purely mental stuff can be translated into our everyday language, in which case its purpose is lost; or if the mental stuff cannot be translated into our everyday language, then it are irrelevant to the discussion, dropping out like a boxed beetle. — Banno

Interesting thread Banno. Just to set aside any misunderstanding, I do not think that conceptual schemes inhere in the mind without or prior to language. To quite the contrary...

Conceptual schemes are metacognitive guidelines of sorts. They are existentially dependent upon rather complex language use. If they alone determine what sorts of things we say are real, then they are a standard for use of the notion. So, rather than thinking that conceptual schemes inhere in the mind without or prior to language, it seems to me that some of the basic elemental constituents of conceptual schemes do, namely rudimentary level thought and belief and all that that requires.

I am in near complete agreement with you and Davidson here, if I understand correctly.



I've recently used the term "incommensurate" to characterize the relationship between two differing views, particularly when there are fundamental differences in frameworks/taxonomies such that they do not translate one into the other as far as certain key terms go. Simply put... one term... more than one referent. The point, I think being made by Davidson and yourself, is that by virtue of my being able to understand that much, by virtue of being able to discern the different referents, the two schemes are translatable one into the other.

Moreover, if there is but one world, this would have to be the case.

So is it nothing or is it objects that make our sentences true? This seems to be a significant inconsistency. — ZzzoneiroCosm

He's equivocating the use of the word make. It's one of the many jokes he scatters through the text, like the use of the phrase "true to the facts" at the start of that paragraph - a reference to another article of his, in which he appears to accept a correspondence theory of truth. I think it clear from other writings, and from secondary sources, that he rejects any correspondence theory of truth. Further while truth for him involves coherence, he rejects coherence theories of truth.

SO no thing makes a statement true; that is, there is no depth in being true, nothing to be explained, no correspondence to facts or what ever. Quite deflating.

Davidson italicises the thing I think in order to emphasis something like that there can be no theory that sets out the things that make a sentence true. I've italicised makes for you, Zzz, for much the same reason. "The point is put better without mentioning facts"

That's all a bit wobbly. With some reservation I might describe Davidson as saying that there can be no theory of what true is - "nothing makes our sentences true"; but that we are caused to assent to some sentences and not others - the things in the world make us believe such-and-such. The bit that worries you is a play on this.

And that brings out the relation between being true and being believed.

Pretty much all that can be said of truth is found in Convention T; any sentence can be put into the form "p" is true if and only if p.

Notice that the "p" on the left is quoted, not used - I have had so many arguments with folk on these fora simply because they didn't...