the question ought to be asked, where do we acquire the ability to recognise general forms in the first place — Wayfarer

We don't. Mistakes are frequent, there's rarely agreement about edge cases and the boundaries (fuzzy as they are) are redrawn all the time in many cases. I can't think where you would get the impression from that there's some question to answer here. We learn that the word 'white' can be used in some range of cases, we certainly don't recognise some 'essence' of white because if we did we wouldn't be tempted to use the term in edge cases and find ourselves misunderstood, but we do.

You can't say 'we don't recognise general forms'. it is one of those comments that blows itself up, like 'all generalisations are false'.

where do we acquire the ability to recognise general forms in the first place; — Wayfarer

What does a philosopher have to contribute to that question?

You can't say 'we don't recognise general forms'. it is one of those comments that blows itself up, like 'all generalisations are false'. — Wayfarer

Why?

Because if you can't recognise general forms, then you can't make general statements, which your statement 'we don't recognise general forms' is an example of. You have to know what a 'general statement' is, even to deny that you can make them; but if you know what it is, then you have to admit that such statements exist. It's the classic Barber paradox.

Because if you can't recognise general forms, then you can't make general statements, which your statement 'we don't recognise general forms' is an example of. You have to know what a 'general statement' is, even to deny that you can make them; but if you know what it is, then you have to admit that such statements exist. — Wayfarer

I'm not seeing the necessary link between 'recignising' general forms and simply knowing how to use generalising terms in a language game. The fact that I know how to use general terms (like 'we' and 'general terms') does not necessitate, in any way I can see, that I 'recognise' anything at all. Not without begging the question (that there's something there to be recognised).

Understood metaphysically, — Andrew M

You jest? (Forgive my irony failure if so.)

Wasn't Quine briefly gesturing to a nominalist translation of sets-talk in terms of shared naming before admitting sets as entities for the sake of exposition of the standard Platonism? And then wasn't Lazerowitz seeing the gesture as support for his proposal: where possible, and in a spirit of charity, read Platonists as positing universals as a shorthand for shared naming?

It's worth noting also that the characterization of words like nouns and adjectives as denoting single universal essences that a person can recognize is largely a caricature – Austin's 'the meaning of a word' is a great commentary on this. A lot of this puzzling springs from never actually attending to what languages do.

where possible, and in a spirit of charity, read Platonists as positing universals as a shorthand for shared naming — bongo fury

When I first read Plato himself, I thought that was all he was on about, and agreed with much of what I thought he was saying. It’s only reading second hand accounts of what professional scholars say Platonists believe that I grokked the full scale of woo that I guess I was too-charitably interpreting as poetic language or something I guess.

You jest? (Forgive my irony failure if so.) — bongo fury

I was going for metaphor. :-) For the jest, see here...

The broader point is that it is easy to be mislead by language and there are plenty of examples of this in the history of philosophy. The irony is that it can happen to any of us even when we think we're attending to it. The Nominalist, in their attempt to exorcise the Platonist spirits, can end up being a mirror-image or dual of the Platonist because of a deeper framing of the problem that neither side has recognized.

So Lazerowitz's strategy here is to expose the roots of the problem itself, not take a side in the dispute.

In effect, the Platonist envisages another realm (of a less tangible sort) that they frame as part of a dual - the material and immaterial. The Nominalist applies their razor to the immaterial side of that duality (because ghosts, extravagence, etc.), but find they are left with an impoverished material world that provides no resources for solving the problem. So the dispute seems unresolvable and interminable.

But the root problem is not that the Platonist has constructed something out of whole cloth that needs to be excised. It's that they have reallocated some things away from their natural home.

Recognizing that dissolves the Problem of Universals - there's no longer a side that needs to be defended.

Wasn't Quine briefly gesturing to a nominalist translation of sets-talk in terms of shared naming before admitting sets as entities for the sake of exposition of the standard Platonism? And then wasn't Lazerowitz seeing the gesture as support for his proposal: where possible, and in a spirit of charity, read Platonists as positing universals as a shorthand for shared naming? — bongo fury

I think more than that, Lazerowitz claims that what the philosopher does "is concealed from himself as well as others" and even that philosophical views are the vehicles for expressing "unconscious fantasies". At the very least it's a call to self-awareness when we use language.

The broader point is that it is easy to be misled by language and there are plenty of examples of this in the history of philosophy. — Andrew M

You don't say. :meh:

The Nominalist, in their attempt to exorcise the Platonist spirits, can end up being a mirror-image or dual of the Platonist because of a deeper framing of the problem that neither side has recognized. — Andrew M

In the fond imaginings of a third kind of philosopher, yes of course... or, do you have examples of such a mirror symmetry?

Lazerowitz does begin with the same too-easy claim, but then proceeds with a perfectly useful analysis that might as well call itself nominalist, like the Quine piece cited. (I'm still not sure you grasped the point of the quoted extract nor Lazerowitz's point about it.) So, examples of the alleged symmetry are lacking.

The Nominalist applies their razor to the immaterial side of that duality (because ghosts, extravagence, etc.), but finds they are left with an impoverished material world that provides no resources for solving the problem. — Andrew M

Which problem? The "problem" of universals? The modern nominalist exchanges that for a more interesting investigation into all of the implications of shared naming...

foundations of math, psychology of consciousness, theory of reference, theory of learning, logic of induction, semiotics etc — bongo fury

Is the material world supposed lacking in resources for these investigations?

The Nominalist, in their attempt to exorcise the Platonist spirits, can end up being a mirror-image or dual of the Platonist because of a deeper framing of the problem that neither side has recognized.
— Andrew M

In the fond imaginings of a third kind of philosopher, yes of course... or, do you have examples of such a mirror symmetry? — bongo fury

Yes. The Platonist embellishes similarities as (capital-N, entity) Names, the Nominalist reduces similarities to (small-n, paper draft [*]) names. Neither side challenges that reclassification nor sheds any light on similarity.

Lazerowitz does begin with the same too-easy claim, but then proceeds with a perfectly useful analysis that might as well call itself nominalist, like the Quine piece cited. (I'm still not sure you grasped the point of the quoted extract nor Lazerowitz's point about it.) So, examples of the alleged symmetry are lacking. — bongo fury

You can call it nominalist, but are you telling us any more than how you're classifying it? ;-) Quine is reclassifying general terms as names there because it's convenient for the logic he is developing. That's fine but it doesn't tell us anything about the ontology of the world, only about his preference.

What we do know is that in the course of our investigations of the world, we can identify similarities and differences in things. That's the natural home that those terms arise in and by which we then classify things (according to our various purposes). So classification itself depends on a prior notion of similarity and difference.

Which problem? The "problem" of universals? The modern nominalist exchanges that for a more interesting investigation into all of the implications of shared naming... — bongo fury

Lazerowitz's analysis is interesting and informative because he's investigating and forming a hypothesis about what philosophers are doing, not discussing how to classify similarity (per the Problem of Universals). More broadly, an investigation and analysis of how language is used in various contexts is also interesting and informative. But, as Wittgenstein notes (quote below), that is not Nominalism.

foundations of math, psychology of consciousness, theory of reference, theory of learning, logic of induction, semiotics etc
— bongo fury

Is the material world supposed lacking in resources for these investigations? — bongo fury

Per "material", yes, which is one side of a Platonic dualist framing that reiterates the reductionism implicit in Nominalism. Similarity, for Nominalists, reduces to just names. Which precludes even the possibility of investigation.

--

[*]

We are not analyzing a phenomenon (e.g. thought) but a concept (e.g. that of thinking), and therefore the use of a word. So it may look as if what we were doing were Nominalism. Nominalists make the mistake of interpreting all words as names, and so of not really describing their use, but only, so to speak, giving a paper draft on such a description. — L. Wittgenstein, PI §383

or, do you have examples of such a mirror symmetry?
— bongo fury

Yes. The Platonist embellishes similarities as (capital-N, entity) Names, the Nominalist reduces similarities to (small-n, paper draft [*]) names. Neither side challenges that reclassification nor sheds any light on similarity. — Andrew M

Yes, I know you think that outcome is inevitable, but I was wondering where, or if, you were finding any examples.

You can call it nominalist, but are you telling us any more than how you're classifying it? ;-) — Andrew M

I don't think nominalists will tend to deny that calling anything by one name rather than another can tell us more than how we are classifying it.

That's fine but it doesn't tell us anything about the ontology of the world, only about his preference. — Andrew M

About the what of the what, now?? This feeds a suspicion that metaphysics is not being easily given up by some of its supposed critics, who need to disparage nominalism because they would rather not be shown a way out.

What we do know is that in the course of our investigations of the world, we can identify similarities and differences in things. — Andrew M

Sure.

That's the natural home that those terms arise in and by which we then classify things (according to our various purposes). — Andrew M

What is? The course of our...?

So classification itself depends on a prior notion of similarity and difference. — Andrew M

What do you mean "prior"? Formed in the process of shared-naming, as a nominalist tends to assume? Or do you, like the Platonist (e.g. Russell), want to cling to a notion of something more innate?

Lazerowitz's analysis is interesting and informative because he's investigating and forming a hypothesis about what philosophers are doing, — Andrew M

Yes, this is such a seductive trope in philosophy: to be too wise to solve any problems. But, as I still say, his analysis might as well call itself nominalist because his suggested reading of Platonist arguments does suggest constructive solutions.

, not discussing how to classify similarity (per the Problem of Universals). — Andrew M

Good, not doing metaphysics, then, just as the nominalist isn't, either, and neither should you (or Russell).

More broadly, an investigation and analysis of how language is used in various contexts is also interesting and informative. But, as Wittgenstein notes (quote below), that is not Nominalism. — Andrew M

Do you mean to damn the enterprise with faint praise, or rather to identify it with Witty's own project, whilst initiating a terminological squabble? If the latter, then hooray, more support for exchanging metaphysics for,

let's assume we are talking about physical particulars and also about the talking of organisms such as ourselves, about those particulars, and let's be especially careful not to get confused when the two targets of our talk overlap, which they probably often must. — bongo fury

which Wittgensteinians can call linguistic analysis if they prefer.

Per "material", yes, which is one side of a Platonic dualist framing that reiterates the reductionism implicit in Nominalism. — Andrew M

Yuk. Metaphysics. Give it up :brow:

Similarity, for Nominalists, reduces to just names. Which precludes even the possibility of investigation. — Andrew M

No, stop assuming this (Platonist canard). Examples please, of doomed investigations into shared naming. Languages of Art for starters if you have to throw it back (and require examples of superlative investigations into shared naming).

Nominalists make the mistake of interpreting all words as names, and so of not really describing their use, but only, so to speak, giving a paper draft on such a description. — L. Wittgenstein, PI §383

Meh. It's a matter of emphasis. Ludo probably reacting against some of his own earlier assumptions about naming, which are not necessarily those of a modern nominalist. But the point certainly is,

We are not analyzing a phenomenon (e.g. thought) but a concept (e.g. that of thinking), and therefore the use of a word. — L. Wittgenstein, PI §383

The pointing of symbols at things by social animals.

Yes. The Platonist embellishes similarities as (capital-N, entity) Names, the Nominalist reduces similarities to (small-n, paper draft [*]) names. Neither side challenges that reclassification nor sheds any light on similarity.
— Andrew M

Yes, I know you think that outcome is inevitable, but I was wondering where, or if, you were finding any examples. — bongo fury

OK, you specifically mentioned Goodman. From Oxford Reference:

Goodman is associated with an extreme nominalism, or mistrust of any appeal to a notion of the similarity between two things, when this is thought of as independent of our linguistic propensities to apply the same term to them. — Oxford Reference - Quick Reference

Now it seems to me that if two things are not similar independent of language, then applying the same term to them doesn't make them similar.

On the other hand, if two things are similar independent of language, that doesn't imply the existence of a third entity for a language term to denote.

The issue in both cases is that similarity doesn't imply a name at all, whether in a Platonic or Nominal sense.

Do you mean,

Now it seems to me that if two things are [not similar non-similar in a sense of similarity] independent of language, then applying the same term to them doesn't make them similar. — Andrew M

?

To the nominalist ("extreme" :lol: or not) this sounds metaphysical, although possibly redeemable in terms of object- and meta-language. Are you in the habit of saying "F=ma, independent of language"? Would you then mean independent of any language (the talk just got metaphysical but through no fault of nominalism), or just higher-level ones?

If you mean,

Now it seems to me that if two things are [not never] similar independent of language, then applying the same term to them doesn't make them similar. — Andrew M

then of course the nominalist disagrees, and is interested in how language creates a similarity between the things.

On the other hand, if two things are similar independent of language, that doesn't imply the existence of a third entity for a language term to denote. — Andrew M

But it does often coincide with use of a general term applying to both: a shared name (or adjective or verb). Then we are presented (sooner or later) with the opportunity to reinterpret the general term as singular, and with questions about how such a choice affects just what entities (e.g. a third one) are thereby implied. Platonist and nominalist might come down on either side of the choice as expected, but the modern nominalist is often prepared to be agnostic on the matter, since there is no fact about it, and because a singular reading (referring to a collective or whole or essence or quality) might be a shorthand for the general reading (referring distributively to all the individual instances).

Lazerowitz was right to read Quine this way. More substantial implications are drawn out in Languages of Art. But here we are well out of the metaphysical mud.

The issue in both cases is that similarity doesn't imply a name at all, whether in a Platonic or Nominal sense. — Andrew M

However glorious.

Do you mean,

Now it seems to me that if two things are [not similar non-similar in a sense of similarity] independent of language, then applying the same term to them doesn't make them similar.
— Andrew M

?

To the nominalist ("extreme" :lol: or not) this sounds metaphysical, although possibly redeemable in terms of object- and meta-language. Are you in the habit of saying "F=ma, independent of language"? Would you then mean independent of any language (the talk just got metaphysical but through no fault of nominalism), or just higher-level ones? — bongo fury

I meant it as shorthand for "independent of our linguistic propensities to apply the same term to them" as in the original Oxford Reference quote.

And, no, I wouldn't normally say "independent of language" because I understand it as an implicit convention in ordinary communication. That is, that language presupposes a world for language to be about.

For example, would you agree that two brontosaurus dinosaurs were similar in the sense of both having four legs before the emergence of human beings and human language?

If you mean,

Now it seems to me that if two things are [not never] similar independent of language, then applying the same term to them doesn't make them similar.
— Andrew M

then of course the nominalist disagrees, and is interested in how language creates a similarity between the things. — bongo fury

So I'm unclear on how you would make sense of that project. It seems to require rejecting the convention I stated above, but for what purpose?

On the other hand, if two things are similar independent of language, that doesn't imply the existence of a third entity for a language term to denote.
— Andrew M

But it does often coincide with use of a general term applying to both: a shared name (or adjective or verb). Then we are presented (sooner or later) with the opportunity to reinterpret the general term as singular, and with questions about how such a choice affects just what entities (e.g. a third one) are thereby implied. Platonist and nominalist might come down on either side of the choice as expected, but the modern nominalist is often prepared to be agnostic on the matter, since there is no fact about it, and because a singular reading (referring to a collective or whole or essence or quality) might be a shorthand for the general reading (referring distributively to all the individual instances). — bongo fury

I assume there is no empirical fact about it, in the sense of an observable difference. However there may be logical (or absurdity) arguments against one or the other of those choices. For example, the Third Man argument which is an infinite regress argument against Plato's Theory of Forms.

On a slightly related note, I've been reading the SEP article about 'fictionalism in mathematics'.

The main argument for fictionalism proceeds essentially by trying to eliminate all of the alternatives to fictionalism. The argument can be put like this:

1. Mathematical sentences like ‘4 is even’ should be read at face value; that is, they should be read as being of the form ‘Fa’ and, hence, as making straightforward claims about the nature of certain objects; e.g., ‘4 is even’ should be read as making a straightforward claim about the nature of the number 4. But

2. If sentences like ‘4 is even’ should be read at face value, and if moreover they are true, then there must actually exist objects of the kinds that they’re about; for instance, if ‘4 is even’ makes a straightforward claim about the nature of the number 4, and if this sentence is literally true, then there must actually exist such a thing as the number 4. Therefore, from (1) and (2), it follows that

3. If sentences like ‘4 is even’ are true, then there are such things as mathematical objects. But
If there are such things as mathematical objects, then they are abstract objects, i.e., nonspatiotemporal objects; for instance, if there is such a thing as the number 4, then it is an abstract object, not a physical or mental object. But

4. There are no such things as abstract objects.

Notice the whole argument is predicated on the simple assertion that mathematical objects - that is, numbers - can't be said to exist, because they're not spatio-temporal. It's a bald assertion (almost an article of faith, dare one suggest).

IN the intro to the article, it is said that:

abstract objects, platonists tell us, are wholly nonphysical, non-mental, nonspatial, nontemporal, and noncausal. Thus, on this view, the number 3 exists independently of us and our thinking, but it does not exist in space or time, it is not a physical or mental object, and it does not enter into causal relations with other objects.

Now, I dispute that numbers are non-mental. They can only be grasped by an intelligence capable of counting, to be sure, but they're not 'mental' in the sense that they exist independently of whether anyone grasps them or not. Frege believed that number is real in the sense that it is quite independent of thought: thought content exists independently of thinking "in the same way", he said "that a pencil exists independently of grasping it. Thought contents are true and bear their relations to one another (and presumably to what they are about) independently of anyone's thinking these thought contents - "just as a planet, even before anyone saw it, was in interaction with other planets." (Frege on Knowing the Third Realm, Tyler Burge.)

Gödel, likewise, believed that what makes mathematics true is that it's descriptive—not of empirical reality, of course, but of an abstract reality. Mathematical intuition is something analogous to a kind of sense perception. In his essay "What Is Cantor's Continuum Hypothesis?", Gödel wrote that we're not seeing things that just happen to be true, we're seeing things that must be true. (Rebecca Goldstein, Godel and the Nature of Mathematical Truth.)

But according to mathematical fictionalism, there can be no such things, because they have no spatio-temporal location; therefore numbers must be 'useful fictions'.

It seems they are 'fictions' because they are more like stories than stones. Complex numbers were controversial once. Now they are intuitively obvious for those brought up among them.

You're missing the point on a very basic level, so I'll repeat.

The point is not skepticism towards whether any purported metaphysical objects exist, are fictional, etc. The point is that theses the the effect that they don't exist, or are fictional, etc. have the same status as the theses that they do.

That is, that language presupposes a world for language to be about. — Andrew M

Sure. But, a world independent of language?

(the talk just got metaphysical but through no fault of nominalism) — bongo fury

So don't blame me...

For example, would you agree that two brontosaurus dinosaurs were similar in the sense of both having four legs before the emergence of human beings and human language? — Andrew M

But clearly something has gone wrong, as the things that a language (or other symbol system) likens to one another clearly don't have to be contemporaneous with it. So of course we can agree on that. But it doesn't get us any nearer to the chimerical "world without language".

So I'm unclear on how you would make sense of that project. It seems to require rejecting the convention I stated above, — Andrew M

Too right. There won't be any fact of the matter of implicit conventions, of course, but one that seems to me to be just as widely asserted is that language presupposes a world already formed/carved/sorted in the terms of the language. (Don't blame me.)

, but for what purpose? — Andrew M

foundations of math, psychology of consciousness, theory of reference, theory of learning, logic of induction, semiotics etc? [...] plenty of philosophy [...] cheerfully non-metaphysical — bongo fury

And ethics.

I assume there is no empirical fact about it, in the sense of an observable difference. However there may be logical (or absurdity) arguments against one or the other of those choices. For example, the Third Man argument which is an infinite regress argument against Plato's Theory of Forms. — Andrew M

This feeds a suspicion that metaphysics is not being easily given up by some of its supposed critics, who need to disparage nominalism because they would rather not be shown a way out. — bongo fury

The way out is to see that we are social animals who think and talk with symbols, whose wholly fictional connection to things is a matter we have to (and learn successfully to) constantly convince each other we are agreed about. Often we can agree that a word points at an abstraction, and often that is because doing so serves as a shorthand for reference to all of the more concrete instances abstracted from.

The point is that theses the the effect that they don't exist, or are fictional, etc. have the same status as the theses that they do. — Snakes Alive

You're missing the point on a basic level. Were this the case, there would be no need for arguments such as 'mathematical fictionalism'. The whole rationale for that argument, which is the subject of many learned papers, is the fact that numbers appear to be real, which undermines materialist views of what exists.

The way out is to see that we are social animals who think and talk with symbols, whose wholly fictional connection to things is a matter we have to (and learn successfully to) constantly convince each other we are agreed about. — bongo fury

And just like other animals, we build nests.

But clearly something has gone wrong, as the things that a language (or other symbol system) likens to one another clearly don't have to be contemporaneous with it. So of course we can agree on that. But it doesn't get us any nearer to the chimerical "world without language". — bongo fury

Wasn't the world prior to the emergence of life a world without language?

Too right. There won't be any fact of the matter of implicit conventions, of course, but one that seems to me to be just as widely asserted is that language presupposes a world already formed/carved/sorted in the terms of the language. (Don't blame me.) — bongo fury

Not "in the terms of the language". For example, scientific language changed as Newtonian mechanics was superseded by relativity and quantum mechanics, and will presumably continue to change in the future. But the world itself didn't change on account of humans using different language to talk about it.

That convention, both in science and in ordinary communication, has been useful.

The way out is to see that we are social animals who think and talk with symbols, whose wholly fictional connection to things is a matter we have to (and learn successfully to) constantly convince each other we are agreed about. Often we can agree that a word points at an abstraction, and often that is because doing so serves as a shorthand for reference to all of the more concrete instances abstracted from. — bongo fury

Then it seems your position precludes any rational basis for agreement. That is, people can agree on one fiction or another (per their preference), but not on how the world is independent of their agreement.

You're missing the point on a basic level. Were this the case, there would be no need for arguments such as 'mathematical fictionalism'. The whole rationale for that argument, which is the subject of many learned papers, is the fact that numbers appear to be real, which undermines materialist views of what exists. — Wayfarer

What are you talking about? I don't think you're following the discussion.

What are you talking about? — Snakes Alive

There are clear resemblances between the arguments about the reality of number, and the arguments about the reality of universals. The argument I referred to, the 'fictionalist' account of mathematics, is made precisely to counter platonic philosophy of mathematics, which holds that abstract entities - namely, numbers - are real. It is not a big leap from 'the reality of number' to 'the reality of universals', it's really the same kind of general issue. And whether you agree that mathematical objects are real or not, the argument itself is a metaphysical issue. Much of Lazerwitz' paper is about the non-reality of universals, so I presume he would have to hold to something very like mathematical fictionalism on the same grounds. So this is directly relevant to the Lazerwitz' argument.

That is, people can agree on one fiction or another (per their preference), but not on how the world is independent of their agreement. — Andrew M

Bingo. The point about such 'mental objects' as numbers (and universals!) is that they are real independently of what you or I might think about them, but that they're only graspable by reason, as Russell says in his discussion of universals in Problems of Philosophy.

Much of Lazerwitz' paper is about the non-reality of universals, so I presume he would have to hold to something very like mathematical fictionalism on the same grounds. — Wayfarer

That is not what the Lazerowitz paper (whichever one you're referring to) is about, nor is it what's outlined in this thread. And that is not the position he would hold, as you would know if you read the OP carefully, rather than 'presuming.'

The point is not a denial of realism about universals or whatever. The point is that any hypothesis as to their reality, pro or con, is equally metaphysical in the relevant sense, and is a matter of proposing to use words in a certain way. That goes for denying that they are real as well as affirming.

Wasn't the world prior to the emergence of life a world without language? — Andrew M

Depends... Is my garden a world without language? And calling a part of it a tree is correct because it is, independent of language?

There won't be any fact of the matter of implicit conventions, of course, but one that seems to me to be just as widely asserted is that language presupposes a world already formed/carved/sorted in the terms of the language. (Don't blame me.)
— bongo fury

Not "in the terms of the language". For example, scientific language changed as Newtonian mechanics was superseded by relativity and quantum mechanics, and will presumably continue to change in the future. But the world itself didn't change on account of humans using different language to talk about it. — Andrew M

So implicit conventions are a matter of fact? Or do you mean that no one reasonably could, considering your argument, persist in the opinion that a theory was speaking "the language of the universe"?

Not that I'm one of those; my point was that both positions are metaphysical (although possibly redeemable in terms of object- and meta-language), and usually dispensible.

Then it seems your position precludes any rational basis for agreement. That is, people can agree on one fiction or another (per their preference), but not on how the world is independent of their agreement. — Andrew M

But "how the world is, independent of our agreement", though a laudable consideration in some contexts, is metaphysical claptrap in most. Science is on Neurath's boat, remaking it from earlier versions of itself, not from something meta.

The point is not a denial of realism about universals or whatever. The point is that any hypothesis as to their reality, pro or con, is equally metaphysical in the relevant sense, and is a matter of proposing to use words in a certain way. — Snakes Alive

I agree, but reducing 'metaphysical' propositions to 'proposals to use words in a certain way' is itself such a proposal (it's also metaphysical). 'Math is fiction' is already close to this reduction of metaphysics to a surface matter of terminological fashion.

Metaphysics, in other words, is often already meta-metaphysics. Perhaps it's less about getting outside of it than it is about doing it less badly.

Wasn't the world prior to the emergence of life a world without language?
— Andrew M

Depends... Is my garden a world without language? And calling a part of it a tree is correct because it is, independent of language? — bongo fury

The answer doesn't depend on those questions. On conventional use, there was no language prior to the emergence of life.

So implicit conventions are a matter of fact? Or do you mean that no one reasonably could, considering your argument, persist in the opinion that a theory was speaking "the language of the universe"?

Not that I'm one of those; my point was that both positions are metaphysical (although possibly redeemable in terms of object- and meta-language), and usually dispensible. — bongo fury

Conventions, like road rules, could be different from what they are. A person doesn't have to follow conventions, but the question might arise as to whether they are saying something useful or merely confusing themselves and others.

Any theory that describes the universe is going to depend on human language. There's no implication that the universe itself would depend on human language.

But "how the world is, independent of our agreement", though a laudable consideration in some contexts, is metaphysical claptrap in most. Science is on Neurath's boat, remaking it from earlier versions of itself, not from something meta. — bongo fury

Neurath's boat works fine as a metaphor for how we investigate the world from within it, and how language evolves as we learn more. Again there's nothing there that implies that the world being investigated depends on our language (or our agreement) about it.

The answer doesn't depend on those questions. — Andrew M

I think it might.

On conventional use, there was no language prior to the emergence of life. — Andrew M

Well I failed to clock that you might have shifted your example from the Jurassic period to some pre-life (Hadean) eon. Was this deliberate? Would you rather talk the similarity of inorganic rocks than of animals? Fine with me. I have a rockery, with no language inside it. I'm happy to say that any similarity between any two parts of it is relative to the language used (from outside) to label the parts. I recognise the notion (of similarity) as meta to any physical or mechanical concepts. Maybe that is a sticking point, I don't know. Perhaps if we clarify the example we may find out.

Any theory that describes the universe is going to depend on human language. There's no implication that the universe itself would depend on human language. — Andrew M

Some important-seeming questions of the 'globalising' variety will always arise. The trick is to be prepared to recognise when one's efforts have developed the symptoms described in the OP, and to then have the humility (or strategic sense) to retreat to more solid ground. — bongo fury

Like, my back garden.

Neurath's boat works fine as a metaphor for how we investigate the world from within it, — Andrew M

Whereas... ?