[If you’re just joining us, please click on the cover icon on the right side of the page to see the post that gathers all the discussions of Braver Reading Group, or click here]
1 DEFINING REALISM
This chapter really just sets the stage, and I don’t have that much interesting to say about it. In the spirit of these posts being helpful for a study guide, I’ll: (1) give the definitions of the Realist Theses Braver considers, (2) give the logical relations between them that he discusses, (3) mention some other realist theses in the literature, and (4) raise a couple of minor issues that may come up again as we move further into the book.
[15] R1 Independence: “The World consists of some fixed totality of mind independent objects” (Putnam, 1981, 49)
[15-17] R2 Correspondence: “Truth involves some sort of correspondence relation between words or thought-sings and external things and sets of things” (Putnam 1981, 49).
[17-19] R3 Uniqueness: “There is exactly one true and complete description of ‘the way the world is’” (Putnam 1981, 49).
[21-21] R4 Bivalence: “The primary tenet of realism, as applied to some given class of statements, is that each statement in the class is determined as true or not true, independently of our knowledge, by some objective reality whose existence and constitution is, again, independent of our knowledge” (Dummett 1981, 434).
[21-23] R5 Passive Knower: “If, whenever I have to make a judgement, I restrain my will s that it extends to what the intellect clearly and distinctly reveals, and no further, then it is quite impossible for me to go wrong” (Descartes, PWD 2:43).
[Chapter 2] R6 Realism of the Subject: “In order that as a science metaphysics may be entitled to claim, not mere fallacious plausibility, but insight and conviction, a critique of reason must itself exhibit the whole stock of a priori concepts, their division according to their various sources (sensibility, understanding, and reason), together with a complete table of them. . . . Metaphysics alone can . . . be brought to such completion and fixity as to require no further change or be capable of any augmentation by new discoveries” (Kant PFM 105/365, 106/366).
2. Logical Relations between the Theses
[16-17] R1 Independence and R2 Correspondence go together pretty naturally.
[17] R1 Independence and R2 Correspondence jointly entail R3 Uniqueness.
[21] R1 Independence, R2 Correspondence, and R3 Uniqueness jointly entail R4 Bivalence (from this and the previous it follows that R1 Independence and R2 Correspondence jointly entail Bivalence)
[21] R1 Bivalence entails R3 Uniqueness.
[22] Sort of suggests that the denial of R5 Passive Knower entails the denial of R2 Correspondence.
[25] R5 Passive Knower again connected to R2 Correspondence and R1 Independence.
3. Other Realist Theses in the Literature:
3.1. Graham Harman’s R7-
The human/world relation is just a special case of the relation between any two entities whatsoever (from his review of the book and brief note at http://larvalsubjects.wordpress.com/2009/05/16/realism-through-the-eyes-of-anti-realism/#comments). This is contrasted with A7 which would say that the human-world relationship is the ground of all the others.
3.2 Dummett’s Epistemic Modesty-
There are unknowable truths. (Braver takes R1 Independence to entail this. It should also be noted that, as far as I can make out, Braver’s Empirical Directive from Hume and Kant just is the anti-realist thesis of Epistemic Hubris, or Verificationism.)
3.3 David Wiggins’ Five Marks of Truth-
These are from the paper “Truth, Invention, and the Meaning of Life” Proceedings of the British Academy; they are discussed in Michael Luntly’s, Contemporary Philosophy of Thought: Truth, World, Content (Wiley Blackwell, 1999).
3.3.1. Truth is the primary dimension of assessment for beliefs and for sentences that express them.
3.3.2. If x is true, then under favorable circumstances x will command convergence. (This clearly follows from some subset of Braver’s theses.)
3.3.3. For all x, where ‘x’ is a sentence expressing a belief, if x is true, then x has content, and it’s truth does not consist in x’s being believed, being hoped for, being wished for, etc. (Connected to Braver’s R5 Passive Knower).
3.3.4. Every true belief is true in virtue of something. (Braver’s R2 Correspondence)
3.3.5.If x1 is true and x2 is true, then x1 & x2 is true; all truths are compatible. (Arguably equivalent to Braver’s R3 Uniqueness)
3.4 Crispin Wright’s- Cognitive Command and Wide Cosmological Role.
These are from Truth and Objectivity (Harvard, 1992). In it Wright argues that there is a “minimal” sense of correspondence (Braver’s R2 Wiggins’ 4th Mark) that everyone can agree with, so that one needs to spell out what makes philosophically loaded senses of correspondence philosophically loaded. He also argues that convergence (Wiggins’ 2nd Mark) might be hold for reasons that have nothing to do with objectivity, such as social pressure to believe a certain way. This issue will certainly come to the fore in Braver’s chapter on Foucault.
Wright proposes that what differentiates robust correspondence from minimal correspondence and (as far as I understand him) objective/Socratic convergence from non-objective/Euthyphronic convergence is whether the following two claims hold of a truth predicate. These are a mouthful. For a clear handout with my take on Wright and realism/anti-realism go HERE . The handout explains the following, among other cool related things).
3.4.1 Cognitive Command- “A discourse exhibits Cognitive Command if, and only if, it is a priori that differences in opinion arising within it can be satisfactorily explained only in terms of “divergent input”, that is, the disputants’ working on the basis of different information (and hence guilty of ignorance or error, depending on the status of that information), or “unsuitable conditions” (resulting in attention or distraction and so in inferential error, or oversight of data and so on), or upward or downwards, or dogma, or failings in other categories already listed).”
3.4.2 Wide Cosmological Role- “Let the width of cosmological role of the subject matter of a discourse be measured by the extent to which citing the kinds of states of affairs with which it deals is potentially contributative to the explanation of things other than, or other than via, our being in attitudinal states which take such states of such affairs as object. . . [the issue is] not whether a class of states of affairs feature in the best explanation of our beliefs about them, but of what else there is, other than our beliefs, of which the citation of such states of affairs can feature in good enough explanations. (196-197)
4. A Couple of Issues
4.1. Bivalence versus the Law of Non-Contradiction
As a rule of logic, the law of bivalence just says that all propositions are such that they are true or false. So stated, this is consistent with the failure of the law of non-contradiction, which states that no proposition is both true and false. That is, bivalence does not say “every statement is exactly one of the pair, true or false.” This is stronger as it requires the law of non-contradiction to prohibit claims from being both.
It is not that Braver is making an elementary logical error. He’s just working with the background assumption that the law of non-contradiction should not be put on the table in these debates. But given how similar Kant’s antinomies and the affection argument against transcendental idealism are to strict paradoxes such as the liar paradox and Russell’s paradox (Graham Priest shows this in Beyond the Limits of Thought), I do think the law of non-contradiction should be on the table.
Take the issue of R3 Uniqueness. Uniqueness typically fails when there is more than one conceptual scheme or whatnot that are equally good. If you follow Goodman (and Kuhn in some voices) and think that there is no sense to the same content being organized by different schemes, then you get genuinely different worlds.
Braver takes this to be a failure in bivalence, on the assumption that if some proposition is true in one admissible world and false in another admissible world then it’s neither true nor false. But it could be a failure in the law of the excluded middle. Maybe being true in two admissible worlds means that the claim is both true and false?
There is actually a debate in the literature between gap (no truth value) and (more than one truth value) glut approaches to super-valuational semantics for vagueness that concerns this exact issue.
This does not mean that I think Bivalence is a law of logic (see- J. Cogburn, “The Logic of Logical Revision: Formalizing Dummett’s Argument,” The Australasian Journal of Philosophy, 83.1 (2005), pp. 15-32.). I just think we need to keep it distinct from the law of excluded middle. My next concern illustrates one reason why.
4.2. One Problem for Claimed Failures of R3 Uniqueness
Typically in science when there are two genuinely incommensurable theories, one or both has such explanatory deficiencies that no-one is likely to think that there is a challenge to uniqueness. In the one case where there were no deficiencies the two theories were proven mathematically inter-derivable (the wave and matrix equation versions of quantum mechanics). Every scientist I know says that what was proven was that there was one theory all along!
The problem is finding genuine cases of non-trivially incommensurable theories that both have equal right to be considered true. (1) Orgone theory and Relativity theory are non-trivially incommensurable, but don’t have equal right to be considered true. (2) The Wave and Matrix equation formulations of quantum physics have equal right to be considered true, but they are not really incommensurable in a non-trivial sense (and hence are the same theory just differently formulated).
When you look at things like ethics and it seems that our best theories in the limits of investigations might still disagree about important things. Examples might include: the amount of our obligation to the poor, or whether lying always involves violating a moral duty. If two theories really did disagree on what our obligations are and they have equal right to be considered true theories (say the best humanity could come up with in the limit of investigation), then my intuition would be that there is still one unique theory of the way the world is, but that it just contains true contradictions (see previous note about law of excluded middle).
4.3. Modality and the denial of R5 Passive Knower and R3 Uniqueness
Finally I worry in a lot of these debates that we end up following Hume and unduly characterizing what counts as a world. If a world is just thought of as a temporal progression of distributions of static property possessions across space, then all sorts of anti-realisms follow.
But once you put modal properties such as possibility and necessity in the world, a lot of supposed anti-realisms aren’t really anti-realisms but just different accounts of the modal nature of reality. Hume couldn’t make sense of the modal notions behind a robust sense of causality, and this led Kant to say causality was just a category shaping our experience. Yet at the same time the transcendental subject applying the categories really looks like it’s engaged in something causal. When you look at the whole thing “sideways” as McDowell puts it, you just have the same old world with some really surprising causal properties.
That is, for all of the neo-Kantian positions you’ve got some human oriented thing, paradigmatically some combination of culture, history, or the subject responsible for different versions of reality. But why doesn’t reality include the makers?
Here’s a better example. Pigeons have a different number of rods and cones in their eyes than do humans. The temptation is to either posit an intervening layer of sense data that is systematically different for Pigeons and humans or get all Goodmanesque and say that pigeons and humans “live in different realities.” But you don’t need to do either of these things. Maybe the ball really is red in itself, but red is just a color that only humans are equipped to detect (and we don’t have a vocabulary for all of the colors the pigeons detect). The real world instantiates all these properties but in a complicated and modally rich way. It’s only if you like Hume and Kant deny modal facts to the real world that you end up saying we live in different worlds (or even worse that we’re really looking at movie theaters inside our own heads).
So whenever one of these thinkers denies R5 Passive Knower as a way to get to the denial of R3 Uniqueness, it always puzzles me. If they use enough causal talk to make themselves clear, then they just seem to be describing one world with a different kind of causal richness than we’d suspected.
This is one of the reasons it is so important not to look at neo-Kantian positions in this “sideways” manner. The danger is you have reinstated all the stuff you were supposed to be wary of (i.e. modal properties like causality being part of the world) just in a weirder, less plausible way. In two chapters, Hegel will come in to try to clean all this up, but first we should get clearer about the the theses themselves, then go through the Kantian moment with Mikhail at the helm. I hope the above is enough to get some conversation started.
Perverse Egalitarianism 
Pingback: A Thing Of This World Reading Group. « Perverse Egalitarianism
Thanks for posting this, wonderful.
Just to echo Kvond, this is a very nice beginning. Thanks.
I do have a few rather basic questions, though.
First off, given R1, R2, and R3, is realism not totally committed to some form of reductionism? I mean, insofar as ‘world’ = a fixed set of independent objects, it seems to follow that this set would have to be denumerable (how else would it be fixed?). If it’s denumerable, then other sets of denumerable claims can be mapped onto it (correspondence, though not in the mathy sense of 1-1 and onto). And, since every (sound) scientific theory is a finite set of true propositions (and therefore countable), how can we affirm R1 and R2, so as to entail R3, and avoid reducing every knowledge claim to this ‘fixed set of mind independent objects’ (which now become R3)? If this is false (i.e. R1 doesn’t entail a denumerable set of ontologically primitive entities), what is the sense of ‘fixed totality’? Second, I’m not sure how R3 really differs from R1, since R3 seems to be something like a recursive definition for the possible concatenations of elements contained in R1. Does this make sense?
It also strikes me that arguments for convergence introduce a related concern, since to say that
seems to mean that true statements concerning the furniture of the world are or will be consistent with one another (i.e. their Sinne may differ but their Beduetungen are identical, and the identity of the latter guarantee the consistency of the former). I take it that that’s the upshot of section 4.2. (and the wave-matrix example from quantum theory) But such a claim is not obviously true. I’m no scientist, or even a very knowledgeable layman, So I may be wrong. But most cases for convergence are mathematical — they involve structural or functional isomorphisms (the quantum theory example is one such case, the math for string theory another). But the math is a tool, not an ontology. So examples of mathematical convergence don’t seem to make a strong enough case. what we seem to need in order for convergence ot be taken seriously (and hence for R3 to be motivated) is an example of a Macro-theory converging with a micro-theory (Quantum physics and Relativity being the perennial favourite here). Simply put, for convergence to be interesting it needs to be an intertheoretic, not intratheoretic (so more like Newtonian physics and Relativity, rather than the math of string theory, or the competing interpretations of quantum thoery). But if that’s the criterion, I’m not sure how many cases of convergence are actually out there, or how much stock one can reasonably put in R3.
I don’t know how clear I’ve been in all this, but hopefully folks can make some sense of it.
Alexei,
These are really interesting observations. What follows is not an attempt to provide a definitive answer.
(1) Point on denumerability-
If you consider the elements of the power set of a X (the set of subsets of X) to be new objects then by standard set theory there is no fixed number of objects, as a powerset is always bigger than the set it is formed from. Standard set theory cops out of this by taking some hierarchy of sets V and not letting you consider V itself to be a set (so you can’t get a powerset of V).
Consider a simplified version of this. Take the natural numbers and adopt axioms that allow the powerset of the set of natural numbers to exists but as a proper class (not itself a set). Then you would have a fixed number of objects in the universe that is non-denumerable (it has the cardinality of the powerset of the natural numbers).
Graham Priest argues that this sort of thing makes standard set theory a cop-out. However, his solution is to consider set theory to involve true contradictions (since the proper class move is in reaction to Russell’s Paradox, Priest ends up saying that the conclusion to Russell’s paradox is both true and false). So your worry ties very closely to my worry about the law of non-contradiction.
(2) I think in way your thoughts on convergence are actually more interesting than what Wiggins has in mind. Wiggins just means that any disagreement on some proposition will converge in the sense that in some Peircean limit of investigation one of the two sides will yield to the other. Wright says that this could happen for rational reasons or maybe for reasons that have nothing to do with rationality (say that one party tortured the other until the other agreed), and that because of this Wiggins convergence would have to presuppose another notion of objectivity to differentiate Socratic convergence (convergence in opinion due to following rational procedures of investigation) from Euthyphronic convergence (convergence for irrational reasons).
Your examples are more interesting than the sense of convergence Wiggins and Wright talk about (though there are implications for that sense) and really do cause a problem for the stuff I said about uniqueness. If I understand correctly, your examples concern the way the solutions to two theories converge in a limit. So the closer (v/c)2 gets to the limit of zero the closer the Relativistic predictions (in some equations) get to the Newtonian predictions, until they are identical. This “reducibility in the limit” shows the manner in which the newer theory is closer to the truth than the older one. Typically systems with the physical magnitudes further away from the limit are the places the older theory went dreadfully wrong. But the reducibility in the limit shows that the new theory also gets right what the older theory got right.
Quick note: it doesn’t always work this prettily, because many cases you get convergence at the infinitary limit but no smooth convergence as you get closer to the limit. See Batterman’s brief account of this in http://plato.stanford.edu/entries/physics-interrelate/
and his excellent book cited therein. This causes a huge problem with the claim that the previous theories are totally superceded. You *have* to use the older theory around some magnitudes. This happens with both with respect to ray theory and classical mechanics (verses wave theory and quantum mechanics). Batterman actually argues that there is a third theory called “semi-classical” mechanics that is not reducible to either.
Even if we get around what Batterman calls attention to , Alexi’s point holds. As far as I understand things (and I don’t that well) these kinds of results hold within specified domains. Alexi is arguing that for reducibility in the limit to motivate R3 Uniqueness we would need some idea that we are on the way to a successor theory that would reduce in limits to both quantum physics and relativity theory. But nobody has anything like that. Instead we have good theories in separate domains that reduce in limits to superceded explanations of those domains.
And now we see the tie to the cruder Wiggins point. If the quantum universe truly contradicts the relativistic universe (here’s where my physics is crap, I don’t know if this is a plausible reconstruction of the nature of the tension between the two!) then Wiggins would say that if the physics truth predicate possesses all the realist marks of truth then any of the disagreements would “converge” in the sense that one or the other party would win.
But if we had convergence in the reducibility in the limit sense between a successor theory equal to both quantum and relativistic domains, this is exactly what you would expect. But then is Alexi’s point that we need to seriously think about the possibility that we won’t get such a thing? And maybe not just because of our epistemic limitations, but because the universe isn’t built that way?
(3) One thing you should not point out to disprove Wiggins is the fact that plenty of previous disagreements did not experience convergence towards one of the sides because we later came to see the whole debate as mistaken. Imagine two people who believe in phlogiston arguing about it’s exact nature. Wiggins would say that the very fact that there was no convergence towards one of the disputants, but rather a rational decision to stop using the category ‘phlogiston’, shows that one of the realist marks of truth for phlogiston discourse was lacking. But that’s not a crazy thing to say.
I’m not seeing how any of the “logical relations” listed here are supposed to work. Unless Braver adds an awful lot of details that you’ve omitted (I haven’t read the book), the relations here just needn’t hold.
I don’t see why R1 and R2 imply R3. You could have multiple equally good ways of describing the world, each of which gives you statements which stand or do not stand in a “correspondence relation” to the fixed block of objects that constitute the world.
I also don’t see why R1 and R2 should go together. One could easily reject the idea that “correspondence relation” is a useful way of thinking of truth (due to the slingshot argument perhaps) while still thinking that what there is is “mind-independent” and eternally fixed. And vice-versa, one could have a non-fixed totality of objects, or (at least some) objects which are “mind-dependent” in some sense, while thinking that truth has to be a two-placed predicate relating statements to facts (or something like that — pick your favorite “correspondence theory of truth”).
I also don’t see how you get from R1, R2, and R3 to R4. It seems to me that you could deny bivalence while saying that there is One True Way The World Is; other ways of talking might fail to be either true or false. R2 doesn’t say anything about what being false amounts to, so the path is open to distinguishing two ways of not being true: being false, and being neither true nor false.
I think R4 can’t imply R3; for instance, Quine is committed to bivalence, but the indeterminacy of translation means there is no “way the world is” — no one uniquely correct description of how things stand. (I think the indeterminacy of translation, and not the possibility of incommensurable conceptual schemes, is the interesting reason to reject R3. Underdetermination is not very interesting; indeterminacy is very interesting.)
Why is R5, the thesis whose substance is “what I clearly and distinctively conceive I can’t be wrong about”, labelled “passive knower”? And why would the rejection of this thesis have anything to do with rejecting a correspondence analysis of truth??
On R7: What is “the” relation between any two entities whatsoever? Everything stands in all sorts of relations to other things; I don’t know what “the” relation is supposed to be. (I can imagine the sort of thing A7 is supposed to be. R7 is not just the negation of A7, I take it. Some relations not being derivative of the human-world relation (whichever one you please, there are lots of relations between the two!), does not commit you to “the” human-world relation (whichever that is) being a “special case” of any higher genera of relation. It can be irreducible without being the ur-ground of all relations.)
Daniel,
Some of these raise interesting points, but some come from not reading Braver’s discussion. I can only do a bit to help:
(1) You write:
“R1 and R2 imply R3. You could have multiple equally good ways of describing the world, each of which gives you statements which stand or do not stand in a “correspondence relation” to the fixed block of objects that constitute the world.”
The thought that all but one of the isn’t really corresponding. If it’s *really corresponding* then the subsentential units are picking out the correct block of objects. But then you get uniqueness.
(2) You write:
“I also don’t see why R1 and R2 should go together.”
Braver didn’t say they entail one another, just that positions that have them together have a certain naturalness, e.g. “If the world is out there with a determinate, independent structure, then it would be odd not to define truth as capturing that structure even if this turned out to be unattainable.” Maybe you don’t find the reaction to the slingshot argument/paradox you put forward odd. I think the vast majority of people who know about the argument would agree with Braver on this one.
But this may just be because I find Wright’s inflation of deflationism to be thoroughly convincing. But so do a lot of other people, so Braver’s point is still correct about the oddity of the view.
(3) You write, “I also don’t see how you get from R1, R2, and R3 to R4. It seems to me that you could deny bivalence while saying that there is One True Way The World Is; other ways of talking might fail to be either true or false. R2 doesn’t say anything about what being false amounts to, so the path is open to distinguishing two ways of not being true: being false, and being neither true nor false.”
You are right. The argument looks like this.
1. Take some arbitrary P.
2. By R 2 correspondence P is true if and only if it corresponds to reality.
3. From the previous line it follows that ~(P is true) if and only if P does not correspond to reality.
4. Since reality is R1 independent and R2 unique, there will be facts of the matter that determine whether P corresponds to reality.
5. So either P corresponds to reality or it does not.
6. But then from lines 5,2,3 P is true or ~(P is true)
So instead of Bivalence, we have at best trivalence.
But is there any reason at all we should take P’s lacking truth to not be equivalent to P’s falsity in this context? Consider the following proof of Crispin Wrights’
7. ~(P is true)
8. | P (assumption for reductio)
9. | P is true (T Schema)
10. |# 7,9 absurdity introduction
11.~P 8-10 reductio
12.~P is true 11 T Schema
So this and the above proof get you
13. P is true or ~P is true 6, 12 logic
What more do you want? You can’t spiel “~P is true” as a case of “neither.”
So the objection fails and Braver gets from R1-R3 to R4 just fine.
Look, it’s not enough to say that in virtue of some bizarro truth predicate that doesn’t satisfy convention T the things don’t follow. If you are going to do that the burden of proof is on you to have an actual example with content.
(4) You write, “I think R4 can’t imply R3; for instance,”
Yep. I overstated Braver’s point. He actually writes,
“Dummett thus defines realism as the commitment to bivalence independento of our abilitity to verify or qcquire sufficient evidence, and this doctrine connects to R3, that there is one true account of the world. The Book of All True Sentences has already been written and all possible sentences are either in it or not; the most we can hope for is to transcribe it. Our ability to determine the truth or falsity of sentences is wholly irrelevant here; truth is radically non-epistemic. . .” 21
Braver was talking about Bivalence plus Dummett’s epistemic account of truth, not Bivalence alone.
(As a parenthetical though, while I agree with you that people make way too much of supposed underdetermination cases, I don’t find indeterminacy in the Quinean context very interesting. I think Gareth Evans pretty decisively showed how compositionality plus a reasonable account of simplicity picks up the slack (“white rabbit” and “white unextended rabbit part” have different stimulus meanings). If you want indeterminacy of meaning, read Mark Wilson’s “Predicate Meets Property” and Stephen Stich’s “Deconstructing the Mind.” They give valid arguments for non-trivial things in Quine’s neighborhood.)
(5) “Why is R5, the thesis whose substance is “what I clearly and distinctively conceive I can’t be wrong about”, labelled “passive knower”? And why would the rejection of this thesis have anything to do with rejecting a correspondence analysis of truth??”
Answer- First Question- The entire early modern tradition in philosophy. Second Question- Kant and the entire continental anti-realist tradition that followed him. On this second part, if you read Braver’s book you will be convinced on this point at least.
(6) Harman’s R7. Note that the corresponding A Theses are not just the negations of the R theses, but rather specific things incompatible with the corresponding R theses that arose while rejecting the R thesis in question.
Harman is contrasting two explanatory strategies, (A7) the one you are stuck with if you are some variety of idealist and (R7) the one you should avail yourself of if you don’t want to end up being some variety of idealist.
Anyhow, I’m sorry if there are typos above. I’m too tired to correct them now. Thanks for getting me on misquoting Harman about the relation between Bivalence and Uniqueness.
First of all, let me say that this reading group experiment is a great idea and that the subject matter, at least for me, could not have come at a better time as I have recently come across Braver’s book and enjoyed reading it. So thank you.
However, I do take issue with some of Braver’s theses of realism. First, he does not take into consideration how one can be a realist/anti-realist in certain domains and not others. For instance, while I may be a metaphysical realist I may also be a moral anti-realist. Braver’s main concern is over metaphysical realism but in later chapters this confusion becomes apparent. Therefore, Foucault can deny the existence of an independent class of things known as mental illnesses or Derrida can deny the existence of Platonic meanings without denying the existence of an independent world. While this does not mean that those thinkers do not deny metaphysical realism (as is more evident for Foucault than Derrida), I do want to say that it is false to infer from the anti-realism in one case to an anti-metaphysical realism in general. Most people are anti-realists about something so it would be uninteresting to be able to find anti-realism in continental philosophy. What is interesting is to find that what continental philosophers have in common is metaphysical anti-realism.
(R2) Correspondence theory is a tough one. I also think that it is a natural fit with R1 but the latter does not entail the former as Devitt points out in Realism and Truth. The question of metaphysical realism is a separate issue from semantic and epistemological concerns and this may be another instance of my above point. But there are realist conceptions of truth (Alston), a denial of which could also be shared by continental philosophers. However, as Alston points out, a realist conception of truth is compatible with metaphysical realism and idealism. So a denial of one does not necessarily involve the denial of the other. Both can be denied but if Braver argues that evidence for the denial of R1 is the undermining of R2, then we have to question this move.
Next, (R4) Bivalence just doesn’t seem like a central tenet of realism because it seems possible to be a dialetheist (someone who holds that there are true contradictions) while also being a metaphysical realist. As Priest remarks, one can be a semantic or metaphysical dialetheism where contradictions are isolated to language in the former or infect the world as in the latter. If it is possible for the world to be in a contradictory state and that world is how it is independent of our beliefs about it then you have R1 without R4.
So then how does this factor into (R3) Uniqueness? R3 may still hold even though R4 doesn’t because if the world is contradictory then the true description of it would be contradictory. If incommensurability is a way of saying that different descriptions of an object can be given that are contradictory then the possibility of there being true contradictions would undermine the apparent stumbling block, in principle, incommensurable schemes have for R3. So it may seem that we’re living in different worlds but this is only a problem if there must be one unified world. What if the world was actually two equally real, incompatible worlds? Of course this is all highly speculative and the metaphysical realist would have others resources for a response to the anti-realist (incommensurability belies shared reference…) but my point has been to argue that a denial of R4 does not entail the repudiation of R3 or, especially, R1.
(R5) Passive knower seems to be entailed by R1 but how passive and active the subject is is important to denying R1. So Kant can argue for an (A5) Active Knower while still maintaining minimal R1 (what Devitt calls Fig-leaf realism, or anti-realism with a fig-leaf). Even the most staunchly contemporary realist philosophers wouldn’t claim that observers are absolutely passive letting the world do what it wants with us. There is a minimum amount of information processing going on to where experience is somewhat constructed. But to what extent is a major issue and I am nowhere near having a thorough answer to that problem. This is to say that we can’t be careless or quick to judgement with how we label philosophers as either R5 or A5 and how this impacts commitment to R1 or R3.
(R6) Realism of the Subject is interesting because in almost every case that Braver looks at, if not every one, in order to maintain a coherent position, the philosopher has recourse to R6. It seems that to deny both R1 and R6 is just frickin’ crazy.
Overall, Braver might have wanted to quote realists for his theses instead of mainly anti-realists (3 from Putnam, 1 from Dummett, 1 from Descartes, and 1 from Kant) in order to avoid charges of constructing straw men. I also think that Braver overstates the analytic component of his book. He very rarely ever comes back to any analytic realist arguments against anti-realism even when defending the positions of the continental philosophers against hypothetical attacks. While I think the clarity of the exposition will make it easier to understand for those coming from outside the continental tradition (and, let’s face it, even for those who are familiar with it), it looks as if he read a handful of analytic books on realism to write the introduction and then turned away. (And I don’t even know how far he got into those books because he argues as if he never read Devitt’s chapter on constructivism.) But I don’t want to get ahead of the curve here so I’ll shut up for now.
Again, thanks for organizing this and allowing others to comment. It has helped me get my thoughts organized and am looking forward to where the discussion takes us.
A few quick comments re: Ricky.
(1) On different kinds of discourses-
You are right that Braver’s book mostly looks at “global anti-realism” as opposed to having different realist or anti-realist takes in different discourses. But a few points are in order.
(1a) This very trait is problematic. Take Wright’s two marks of Cognitive Command. Clearly some judgments in literary theory satisfy this, if someone claims that Hamlet is really a space alien they are cognitively deficient in some way. But if two people disagree about whether Hamlet has sexual feelings towards his mom, neither one of them might be cognitively deficient. Likewise for linguistics. If two people disagree about whether quantificational structures give rise to ambiguities, one of the parties is missing some crucial bit of information or being irrational. But two rational people can disagree about whether deep structure exists.
Many of the distinctions that supposedly mark off realist discourses actually apply *within* any discourse to mark of judements or particular concepts. Wright’s Truth and Objectivity is perhaps the greatest book that never seems to be cognizant of this.
This is related to the fact that we make valid and invalid arguments using predicates from different discourses in the same argument. If these predicates systematically get different logics the way Dummett and Wright entertain it becomes very difficult to understand such cross discourse references (this is just a generalization of the Frege Geach problem for the non-cognitivist claim that some discourse lacks a truth value, and part of the reason that Wright argues for minimal truth).
(1b) In any case, Braver does present some philosophers of having different kinds of realism for different subject matters. Kant’s views are different for phenomena, noumena, and transcendental subject. Hegel’s views are different at the consciousness, self-consciousness, and absolute idealism stage. Nietzsche’s views are different for stage 5 and stage 6 physics.
All this being said, the anti-realisms Braver discusses do tend to be “global” in that the philosophers in question don’t parse among things like, one kind for ethics, another for math, another for fundamental sciences, another for special sciences, another for aesthetics, etcetera.
For this to be a good critique of his project, we’d have to see if failing to do so leads to a critical misunderstanding of Kant, Nietzsche, early or late Heidegger, Foucault, or Derrida. Or see if these thinkers aren’t worth taking seriously because they didn’t proceed discourse by discourse. This seems mistaken to me. Most of Dummett, Wright, and McDowell’s actual work isn’t just about any discourse, but really applies just as globally as the continental thinkers in Braver’s treatment. Heidegger did have a philosophy of science, but to understand it we need to understand the global positions in the background, likewise with understanding the interesting particular things Dummett had to say about the past.
(2) Dialetheism does not contradict Bivalence. If a claim is both true and false, then it is still true or false. So the fact that one can be a dialetheist (denying the law of non-contradiction) because of self-referential paradoxes as you get with Priest and still be a realist does not imply anything about bivalence.
The kind of dialetheism I defend in J. Cogburn, “The Philosophical Basis of What? The Anti-Realist Case For Dialethism,” in The Law of Non-Contradiction, ed. Graham Priest, J.C. Beall, and Bradley Armour-Garb, Oxford University Press, (2004) is anti-realist (though I think realist enough!). But neither Priest’s dialetheism, nor my weird anti-realist extension of it, undermines the Dummettian claim that bivalence is a central part of the anti-realism debate.
(3) As far as the analytic component, I’m only chapter 4 but there’s already been a couple of very rich discussions in the text of Sellars, Davidson, Goodman, and Wittgenstein that are interesting in their own rights, and the extended footnotes are goldmines. I think the further place this stuff will tie into that he doesn’t go into with that much depth is John McDowell. But that’s not a criticism of Braver, just a great project his book makes much more doable (clearly from the above I’m very interested in Graham Priest, whom Braver footnotes, in these contexts too).
(4) I don’t remember if I’ve read Devitt’s chapter on constructivism. I might have in graduate school; I’ll check it out.
Dr. Cogburn,
I have been confused on an issue similar to the one Ricky brings up, so perhaps you can help me out.
Okay, so R2 says that truth is the correspondence between thought-propositions and a fixed, external world. It seems then you can deny R5 Passive Knower in at least two different ways. You can say, “there is no passive knower, because our thought-propositions actively structure the external, perceived world.” This is the classic Kantian root. However, it seems as if you could also just deny the literal existence of thought-propositions but maintain R1 at the same time.
Then, you could have R1 without R2, because in denying R2 you aren’t saying there isn’t a fixed, eternal world, you are just denying the literal existence of thoughts that are linked up to a fixed world. Instead, take the Heideggerian root and say that truth is a disclosure relationship to a fixed world, and any R2 correspondence of thoughts and worldwould be derivative and not metaphysically “real.” You would then have R1, A2 and a deflated R2, and A5. But “Active knower” would have to be redefined with “thoughts” no longer being the ultimate criterion for what knowledge is.
Hopefully that makes sense. Essentially I am saying that A5 can go against R2, but only because it denies the reality of “thoughts” and not because of a denial of R1. Do you think that works? It seems like something Braver glosses over, but I am not sure.
Orestes,
That seems right to me. I think it ties to Graham’s R7 and A7 distinction, as well as the kind of debates about eliminativism that Brassier starts his book “Nihil Unbound” with. I’m going to think more about it.
I always find it exciting when someone like Jon can take my relatively incoherent remarks and turn them into something truly interesting. So Jon: thanks for taking the time to unpack the mess of material I jammed together. I’m hoping to follow up a little, if you’re willing. If I’m lucky, the following will also cast more light on my claims concerning convergence.
Concerning denumerability again:
Initially, I did have something like a power set in mind, which is paradigmatically not well-ordered (and hence not necessarily a ‘fixed totality’) if memory serves. but a few things occurred to me. First, it’s simply not clear to me that, besides the Urelemente of R1, I should accept the claim that the members of a given power set are also members of R1 for precisely the standard reasons. Although, in the case of the sciences, such a hierarchy is less of a cop-out than it is an acknowledgment of the stratified character of phenomena.
consider an example: the elements of the periodic table constitute a fixed totality of mind independent objects, complete with discrete and measurable properties. Like your examples concerning natural numbers, this set (call it E) is countable (because it’s finite) even though its power may not be (I don’t know enough chemistry to say whether there is an upper combinatorial limit on molecules, although intuitively there should be some such thing). However, it’s not clear to me that H2O, which is a member of PE, and which has distinct properties form its Urelemente, has the same ‘ontological status’ as the Urelemente of E. Moreover, it seems to me that the set of true (scientific or theoretical) statements about H2O is finitely large, and hence can be ordered and mapped onto E. Point being, there’s some kind of compositional relationship between scientific statements, P(R1) and R1 itself, which mirrors the relationships between the special sciences, their domains of inquiry, and what’s really real. Simply put, P(R1) can be carved at its joints so as to become tractable.
The issue I’m trying to raise, I suppose, has to do with whether it makes sense to claim that R1 must be a fixed totality (instead of, say, a fluctuating one, which would render R1 rather unrealist — a kind of flat ontology, perhaps, which would then have to deal with power sets and the law of non-contradiction), since it seems to commit one to two claims:
Rephrased, then: the price of a fixed totality of objects seems to be (1) reduction (2) a stratification of scientific schemata in order to avoid the explosion of ontological commitments (3) disavowel of flat ontologies in order to maintain the law non-contradiction, and block the paradoxes of naive set theory. But this seems to create problems for R3, since R3 now entails a robust sense of convergence (of the form you outlined above), since it has to be consistent with the reduction to a countable set of ontological primitives, and to the intertheoretic relations among special sciences.
All this to say, then, that my problem concerns how R1 is ordered or fixed and the implications of such a claim (which include, among other things, R3 Uniqueness). If it’s well ordered, it’s denumerable. If it’s denumerable, then there is a unique ordering of elements. Further, although the power set of R1 may not be denumerable (and this depends upon whether R1 is a finite set or merely countable — and it strikes me that to be informative R1 has to be a set of types, and hence finite), one can still construct equivalence classes (or something akin to it) in order to generate countability (special sciences and their respective schemata).
Just a thought, anyway.
I’m not sure we should get into the issue of well-ordering. By the axiom of choice (in first order set theory) every set can be well ordered, including the powerset of the set of natural numbers.- which is genuinely weird and tied to the Tarski-Banach paradox.
But I like what you are doing here quite a bit, and I think your subconclusion is provocative and plausible-
“The price of a fixed totality of objects seems to be (1) reduction (2) a stratification of scientific schemata in order to avoid the explosion of ontological commitments (3) disavowel of flat ontologies in order to maintain the law non-contradiction, and block the paradoxes of naive set theory.”
While there may be some clever way for the R1 person to get out of these, this has got to encapsulate the standard R1 maneuvers.
I’m still not getting the tie of this issue for R1 to R3 Uniqueness, except by way of (2). Is your thought that the ontological novelty of emergent properties causes a problem with the idea that we can state the complete story of nature in a finite enough (maybe some form of denumerability) manner?
I think this will depend a lot on one’s theory of how language is working. If you think that for something to be stateable it has to be decidable in some manner, then I think you are right that R3 is going to go out the window. We’ll have a lot of different decidable theories about the world and no way to get them all together (limitation results such as the unprovability of the halting problem and Goedel’s theorems come in here). But if you think human beings are already good at grasping and describing things that are not algorithmic then you could still have R3. Here the one true and complete description would include a rich vocabulary to discuss phenomena that is not algorithmic.
Does that make sense in this context? I may be pushing your point too much in terms of the issue of something being decidable because that’s something I like to think about emergence in terms of (e.g. I hate that I’m doing this here, but I’ve got a couple of papers considering failures in algorithmicity:
[1] J. Cogburn, “Deconstructing Dummett’s Anti-Realism: A New Argument Against Church’s Thesis,” The Logica Yearbook (2002),
[2] J. Cogburn and M. Silcox, “Computing Machinery and Emergence,” Minds and Machines, 15.1 (2005), pp. 73-89.
[3] M. Silcox and J. Cogburn, “Computability Theory and Literary Competence,” The British Journal of Aesthetics, 46.5, (2006), pp. 369-386.
I’ve got two more with Mark Silcox and Jason Megill (respectively) that concern emergence and failure of algorithmicity.)
Great comments/opening post sequence. I’m going to post a rejoinder on Chapter 1 tomorrow morning, hopefully making sure I don’t repeat some of Jon’s observations.
This post has been getting a lot of hits, so if you are reading this and would like to contribute to the conversation, please do so as we will all benefit from as many angles/questions as possible. Feel free to address any aspects of chapter 1, not just Jon’s observations or commenters’ reaction.
“The thought that all but one of the isn’t really corresponding. If it’s *really corresponding* then the subsentential units are picking out the correct block of objects. But then you get uniqueness.”
So R2 is supposed to be something stronger than “whatever is true is true because it is related to a fact which makes it true” (or something like that). I was thinking you could still have each sentence “made true” by a fact, but not need there to be one privileged vocabulary which can pick out the facts (the facts don’t need to be structured propositionally). But apparently R2 has a semantic atomism built into it. Okay.
“7. ~(P is true)
8. | P (assumption for reductio)
9. | P is true (T Schema)
10. |# 7,9 absurdity introduction
11.~P 8-10 reductio”
This isn’t a move that one should allow if one wants to use a three-valued logic. The reductio move says that the assumption of “P” cannot be right, but then it licenses one to assert “~P”. But if one is using a three-valued logic, then this is too quick. Adding “P” might lead to absurdity because P is neither true nor false (and the same would be the case for ~P). A modified rule for reductio ad absurdum in a three-valued logic would give you “~(P is true)” for 11. (At a glance, I think that should work; I don’t like using natural deduction systems (trees are just easier to work with IMO), but that should line up the natural deduction rule for reductio with the relevant set of moves in constructing a tree. I know the tree rules for K3 work fine, which is why I’m not troubled by not being able to summon up a natural deduction rule here.)
The T-schema biconditional can still work normally in all this, so I don’t see that I need any weird notion of truth here. I just need truth-value gaps (such as “William Tell shot an apple off his son’s head”, which is neither true nor false, at least for Frege, since there was no such man — it’s not false because it’s not any more true that he didn’t shoot an apple off his son’s head than that he did). I’ve never understood how someone could think that gaps are less appealing that gluts, if we must have one or the other.
On the parenthetical about indeterminacy: I don’t think that Quine’s account of “stimulus meaning” makes sense generally; I’m a Davidsonian when it comes to Quineanism. Which also means that I don’t see why people regard the slingshot argument as paradoxical: without the sort of “correspondence theory” that the Slingshot kills, we can still say “Something is true if it says that something is so, and it is so, or if it says something is not so, and it is not so; something is false if it says something is not so, and it is so, or if it says that it is so, and it is not so” (Aristotle’s “theory of truth”). That sort of thing can stand firm without any trouble from the Slingshot. (So I also see no connection between rejecting the “correspondence theory of truth” and being an idealist; I am aware that many people do think the two are connected, and I think that’s just confused. The one does not entail the other. Nor does rejecting the slingshot tell you anything about someone’s epistemological tenor; Frege is as staunch a Platonist and anti-idealist as they come.)
Daniel,
I think the main point is that, as Dummett says, the kind of trivalence you get from a determinate three value logic is still just as realist as a determinate two valued logic (on his new view God uses a three value logic and the rest of us use intuitionist logic). If you are adding gaps for things like presupposition failure then that doesn’t have much to do with the realism/anti-realism issue.
So I think it’s entirely licit for Braver to not worry about it. He could have put a footnote in saying something to the effect that if you accepted a gap solution too presupposition failure, then trivalence is metaphysically realist in that sense (again, Dummett says the same thing). But it wouldn’t effect the realism/anti-realism issue.
You’d need a separate anti-Dummettian argument to show that determinate trivalence has no metaphysical presuppositions. For Dummett, it will end up undermining verificationism (Braver’s “Empirical Directive”) and be realist in that sense.
I just mentioned stimulus meaning because you brought up Quinean indeterminacy. I don’t think Quine’s arguments are valid and agree with you that the framework is unhelpful. That’s why I go with Wilson and Stich (and some Putnam) for interesting valid arguments to the underdetermination of meaning.
The stuff you raise about slingshot is really interesting!
For Braver it’s important to realize that the whole story gets off the ground by Kant’s giving up the notion of a passive subject, and installing the phenomenal/noumenal distinction in terms of that giving up. The stuff about correspondence follows from that, not the other way around. So I think what he’s up to is consistent with your interesting points.
Jon,
Yes, let’s leave the notion of well-ordering aside — I’m already at the limit of my mathematical competence (if not already past it), and the Tarski-Banach paradox is making my head hurt. Any more thought about it will cause my poor noggin to spontaneously rearrange itself into two heads with twice the pain.
This said, everything you’ve written makes total sense to me, especially concerning decidability. So I’ll definitely have to look at your papers on this issue.
Anyway,let me respond to your question. You ask,
Short answer: yes, that’s exactly right. The problem I’m hamfistedly trying to articulate seems to boil down to the following: emergentism is not compatible with reductionism, since — by definition — emergentism means something is more than the sum of its parts, and can’t be reduced to more basic, ontologically primitive elements, without loss. So, if R1 commits us to reduction, then it also forces us to reject all metaphysical accounts of emergence. We either affirm R1 (which entails reduction), or we affirm emergentism (which forces us to reject reduction — and R1 for the sake of consistency). Furthermore, if we choose to affirm some form or emergent account, and do away with reduction, there doesn’t seem to be any way to construct R1. There’s simply no tractable way to to identify ontological primitives on an emergentist account. If, for instance, phenomenon P emerges from the interaction among thingies a, b, and c, do we count 1 primitive entity, 3, 4, or 5? And this is precisely the problem of decidability you bring up. So I suppose I’m emphatically affirming the thesis that to be stateable = to be decidable, along with the claim that this thesis poses problems for R3.
Now that i think about the matter, there may even be a further problem. Were one to attempt to dodge the ‘decidability bullet’ affirming this statement of yours
would they not generate a paradox? If one’s realist ontology requires a metalanguage that is rich enough to capture non-algorithmic phenomena, then this vocabulary must also be ontologically primitive. It needs to be a member of R1. If it is, however, R1 is incoherent, since the fixed totality of mind independent objects includes a mind dependent vocabulary. Again, this is hardly well thought out. I’m just trying to bounce ideas.
Your final point ties to an interesting persistent theme- how does one categorize the representer and representational machinery?
I’m getting this argument from you:
(1) Assume that the fixed totality of objects is a set of objects O which possess some ontologically primitive properties.
(2) But then mereological sums of objects in O that possess ontologically emergent properties, such as “being a hurricane” either (2a) do not count as members of O, or (2b) make O such that it has no determinate number (since characterizing the objects that bear emergent properties will require multiple applications of the powerset object to O.
(3) So either O does not include all of the objects in the universe, or the number of things in O is radically indeterminate.
Then your final point considers what happens when you consider representations (thoughts, words) themselves: are they part of O or not?
A flat ontology is going to try to follow Harman’s R7 thesis, saying that mind-mind and mind-world relations need to be seen as instances of world-world relations and hence put them in O. But then as you note earlier, you are probably going to have to give up on the claim that O has a determinate number of objects in it, and almost certainly then follow Priest and become a dialetheist (both for reasons having to do with self-reference paradoxes and for reasons having to do with theories equally good in the limit of investigation disagreeing on some claims).
I still think if you amended R1 in all these ways you could do this and keep some form of R3 Uniqueness. The one true story at the end of the day describes a world no stable number of objects and also with true contradictions.
I want to argue that- R3 Uniqueness only gets threatened, as you note, if you do not follow Harman’s R7 and try to see mind-mind and mind-world relations as instances of world-world relations. Because then you’ve got this fixed set of real stuff that can be described in all these different ways (emergent properties become secondary properties), so you have to go with Kant and undermine R6 Passive knower in a way that ends up allowing for alternative schemes (underming R3).
I hope that Priest and Harman save the day for the realist though by allowing us to just minimally amend R1.
Wow, I read all comments (again) in one sitting, I think my head is about to explode – great stuff, I wish I was better versed in all of this so that I can at least nod understandingly…
Hopefully I’m not too late to jump in on this discussion. As I read Braver’s book– and it will be difficult for me to keep up this summer due to other obligations –I am struck by a sense of the uncanny. On the one hand, the picture of realism Braver presents in the first chapter strikes me as doing what a lot of the speculative realists have been accused of with respect to anti-realism and correlationism: turning realism into a straw man. Let me hasten to add that I do not necessarily think this is a bad thing. Braver’s service is to render us not a sophisticated picture of realism (Whitehead, a realist, for example, would fit none of these theses), but to give us a robust picture of anti-realist debates among Continental philosophers. I am also struck, as I read Braver’s chapter on Kant, that he attributes all of the claims I have attributed to Kant to Kant. Indeed, in many instances he uses phrases identical to my own in describe Kant’s thought.
All of this aside, I find that I do not fit any of the six theses attributed to realism, so I’m left wondering whether I’m a realist, an anti-realist, or something else besides.
I do not fit R1 because I do not endorse the thesis that the world consists of a fixed totality of mind-independent objects. In other words, I do not advocate the position that the objects of the world are denumerable. I am not of this position because I believe that the set of objects is greater than can be humanly counted, but rather because I hold that new objects are coming into existence all the time. For example, the gorgeous peppers growing in my garden. It seems to me that realism is here being conflated with a vulgar sort of materialism or atomism, where the universe is held to ultimately be composed of a finite number of indivisible atomic units.
R2 does not hold for me either, and is my point of greatest proximity with the anti-realists. Back in the day (January) when we were all having the lively realism wars, some of you might recall that I proposed, among others, two principles: the Ontic Principle and Latour’s Principle or the Principle of Translation. The Ontic Principle states that there is no difference that does not make a difference. When I claim that there is no difference that does not make a difference, I am not making a value statement to the effect that all differences are “important”, but rather am stating that all differences act and produce other differences. The Principle of Translation states that there is no transportation without translation. What I mean by this is that when the difference of one object acts on another object it translates or transforms that difference in a way unique to the receiving object. Thus, for example, my pepper plant “translates” the difference of sunlight producing energy in the form of sugars that it uses to produce its fruit and leaves. The process of translation thus transforms the differences of other objects in a way particular to the object doing the translation.
Since I hold, with Graham Harman, that this principle holds for relations among all beings (and not just the relation between mind and world), it is clear that it is logically impossible for me to advocate a correspondence theory of truth. Why? Because insofar as all objects translate the differences of the objects they receive, there can be no question of an object “mirroring” or representing the world in a one-to-one correspondence relation.
My point of contention with anti-realisms or correlationisms is thus not in the thesis that mind transforms data of the world such that it does not represent the world as it is “in-itself”, but that it includes the human in all relations, rather than generalizing the thesis of translation to all object-object relations. I advocate a thesis that might be described as a “generalized Kantianism of objects”. That is, just as Kant holds that objects conform to mind, I hold that objects conform to whatever object is translating them. In this respect, I advocate a position similar to Leibniz’s perspectivism, where every monad or entelechy (object) represents the world from a particular point of view.
Given the Principle of Translation, it is therefore clear that I cannot advocate the third thesis (R3) either. The only condition under which it would be possible to advocate the thesis that there is a unique description of the world, would be under the condition that objects do not translate the differences of other objects creating new properties. However, since every object translates the difference of other objects in its own unique way, it follows that there cannot be a unique description of the world.
R4 then appears to fall as well, as does R5 since the process of translation is not a passive process, but rather an active transformation of differences.
One thing I’ve noticed in the realism discussions is that realism is often identified with scientism. That is, the classic realist position is that real objects are physical objects. This, for example, seems to be Brassier’s position. I think one fundamental way in which my position differs from Brassier’s on this count is that, since for me the criteria of existence dependence on whether or not something produces a difference, anything that produces a difference has the status of being real. Quarks are real. But then, so is the character of Half-Cock Jack in Neal Stephenson’s novel Quicksilver. The result is that I don’t divide the universe into the real or nature on the one hand, and culture and the human on the other hand. Rather, because both of these domains produce differences, I hold that they are real. As a consequence, a number of the questions revolving around issues of representation just don’t arise within my ontology. Where the person that adopts a representational theory of knowledge is led to ask how a thought (something unreal or “only in the mind”) can correspond to an object (something real), my question becomes a bit different. I instead ask not how one thing can represent another thing, but instead investigate the nature of the relation between two real objects (the thought being one object and the flower, for example, being the other object).
Levi, thanks for your input. I think the issue in the first chapter, as I understood it, was to introduce the discussion of realism as it is present in the literature dedicated to that matter and a set of theses (R1-R6) being a crafty way of putting many positions into one large picture of what people are saying realism is. I think from the comments above I can already tell that not everyone is happy with this particular set of theses and their interaction, but you do raise an interesting point, I think, when you ask whether according to Braver’s theses your position can count as “realist” and I also think that you clearly demonstrate that it does not. The question for me is whether we should change Braver’s presentation of realism to include the elements that you think might be missing in order to qualify your positions as those of “realism” or whether perhaps your positions are not properly realist? I think that looking back at Realism Wars discussions now from a perspective of time and Braver’s succinct presentation of the literature on the matter, I wonder if many of the disagreements and passions could have been avoided if we settled first on what exactly “realism” is and is not.
Having reread chapter 1, I wonder if Braver’s presentation of R1 in “Guide to Matrices” and in Chapter 1 itself is confusing in that in the Guide, he cites Putnam who talks about “fixed totality of objects” while in the Chapter 1 itself we only get a metaphysical position that “set of objects or states of affairs, which does not rely upon us in any way, exists” [15] which would not be problematic if we take new objects to be constantly appearing, as long as their independence is accepted.
Again, I think we need to distinguish between Braver’s presentation of the views on realism that exist “out there” and our own (preferred) understanding of realism. In a sense, it seems like a good place to begin the conversation, but I’m sure that it would not cut it if that was all there was. I take the intention of the book to be what it claims that it is, a demonstration that continental anti-realism is working with these realist theses and there’s a connection therefore between the two philosophical traditions.
Levi,
Thanks, that’s great stuff. I’m going to think of how these connect up with Alexei’s points and also my own weird thoughts on emergence and undecidability.
I tend to think of you and Graham’s views (I just started reading Brassier) as being forms of realism that take seriously the lessons of Kant and Hegel, and in virtue of this end up departing significantly from R1-R5, which Braver shows correspond to Hegel’s portrayal of pre-Kantian realism.
If I understand right, you (object oriented) guys agree with Hegel’s criticism of Kant but reject the characteristic Hegelian anti-noumenal move (for now, the Hegel Move) that Braver brilliantly shows leads to Hegel’s absolute idealism, Nietzsche’s “stage six metaphysics,” and Heidegger’s “phenomenological ontology.”
Braver takes the late Heidegger to reject the Hegel Move in a further “anti-realist” manner (I’m just now getting to that part of the book) that leads to Foucault and Derrida. I’m fascinated by the prospect of being able to clearly contrast that “anti-realist” rejection of the Hegel Move with what I take to be the speculative realist “realist” rejection, and hope that having to post with this group allows me to begin to clarify the contrast.
This all ties deeply to John McDowell too, and think that Braver will allow us to ultimately be able to place his neo-Hegelian anti-anti-realism in the same dialectical space with varieties of speculative realism. Something interesting will happen once the comparisons are made.
Plausible forms of realism to me must: (1) take seriously *all* the problems that led to transcendental idealism (modality, paradoxes of totality/self-reference, normativity, and linguistic and epistemic problems of the external world), (2) not be subject to revised forms of the affection problem that Hegel and Schopenhauer raised for Kant, and (3) reject the Hegel Move in a principled way.
For me this close reading of Braver is a prolegomena to being able to think through (1)-(3) in reference to a set of thinkers (including you!). Over the next few weeks I hope to minimally: (a) get very clear about the “affection problem” Hegel/Schopenhauer raise for transcendental idealism, (2) get very clear about the Hegel Move (of which absolute idealism, stage six physics, and phenomenological ontology are all instances), (3) get very clear about the late Heidegger’s critique of the Hegel Move as leading to increased anti-realism.
Anyhow, I’ll be fascinated to continue to read any of your on these things.
Mikhail,
Excellent points. A lot of the above discussion focuses on the number of objects being fixed. You are right that this is orthogonal to the question of whether the objects themselves depend upon us. It’s neat to see that the latter notion is what is doing the work in much of Braver’s discussion.
In presenting some of the other “realist” theses in the literature by Wiggins, Harman, and Wright I tried to note that the fact that Braver misses them doesn’t counterexemplify anything he’s doing. As you note, he just presents the theses he needs to be able to explain the evolution of certain kinds of anti-realism. Anyhow, you put this much better than I did, so I’m just saying I agree with what you said.
[Brief note that just occurred to me: as far as the “number of things in the world” issue, there are related paradoxes from debates concerning possible worlds; John Divers discusses pretty clearly in his great book called “Possible Worlds.”]
Thanks, Jon – I think as far as the educational aspect of this group is concerned, I am learning much more than I would have liked to already anyway 🙂 I think it’s essential that we situate the discussion in the appropriate presentation of the field (maybe it’s my “book report” continentalism coming through) before we are able to move on. Maybe Lee will address this in his Response eventually. I wonder however what would be an alternative to this simple presentation of the matter as it is already discussed? I see the point of saying “Well, your realism is not what I think realism is (or rather, should be), so your points are of no interest” I’m not saying that this is Levi’s position, I’m just simplifying a possible objection to Braver’s theses. How do we begin to define realism from a blank slate? I think the answer for me is clearly “We cannot” and that is what I mean when I say that Braver’s summary is excellent starting point of this discussion, otherwise it would have been a simple dogmatic imposition or a secondary literature report.
Oops, I just now am reading where late Heidegger does not in Braver’s reading reject what I called “the Hegel Move” above, but rather bites some of the bullets that arises from it (mainly, from what I’ve got thus far, the problem of trying to account for falsity in a world of pure phenomena).
(internet connection = 😦 since yesterday, so this we’ll be exceedingly quick)
Wow Jon, you’ve done an amazing job making me sound as if I have a real argument buried in all of those tangled sentences of mine. Thanks.
So, let’s continue with the fiction: your 3-step argument from set O is precisely what I’m getting at. That’s exactly right. There’s one question I would like to ask, though, concerning your use of mereology and emergentism in the same sentence (step [2]): do you think that mereology entails emergentism? A fusion theory I guess? I’m not an expert, so I might be totally wrong, but the hallmarks of emergentist theories (e.g. downward causation, property supervenience) are usually characterized as non-mereological. Roughly put, if mereology is bottom up, emergentism is top-down. Maybe I’m missing something, but if one has a fusion theory why call it emergentist, since the basic idea is still a bottom up causal interaction (fusion of strata)?
Before continuing, I also want to ask you whether this is a fair list of necessary amendments to R1 (call this ‘amended R1’ SR1) in light of the argument from O:
Is that fair? Would it also be fair to say that we would need an analogous reformulation of R3 (i.e. at the very least include a temporal marker, a complete, unique description at time T)? if it is, what does (2) amount to? (does it simply mean, that the property of mindedness is mind-independent?)
These questions aside, I’m not sure I understand how Harman’s R7 resolves the problem we’re gesturing at, or how we can understand his work in terms of a flat ontology. Surely, claims concerning substance — even perishable substance — generate a hierarchy (e.g. primary and secondary qualities), and surely understanding objects as infinitely withdrawn introduces some level of depth, albeit perhaps on in a catechrestic way.
More concretely though, as long as the set R1 or SR1 =df the totality of mind independent objects, R7 isn’t going to do us much good. For even if all relations are world-world, those which are minded world-world still can’t be a member of any variant of R1 without generating a paradox. For they’re still mind-dependent. The contradiction here is in adjecto, and not logical, semantic, or ‘metaphysical,’ so I doubt that Priest would extend his dialetheist claims to it. Simply, put, I don’t think it matters here what the mind is made up of, or whether the particular relationship between mind and world is simply a special case of a relationship among monads. It’s still mind dependent. And in this case, it strikes me that there’s a flaw in our reasoninig, not a fundamental, metaphysically grounded contradiction
This may be really simple minded, but can we start by following Meillassoux here and reinterpret R1 to be a claim about the world when human beings are subtracted?
Instead of
“The World consists of some fixed totality of mind independent objects”
we get something either temporal (and we could do this in terms of what happens after human extinction),
“Before there were human minds, there was a fixed totality of objects,”
or modal,
“If human beings were to be removed from the universe, there would still be the same fixed totality of non-human objects.”
That’s still a substantial realism thesis. Of course we still have cardinality issues when phrased that way, but then maybe we can follow Mikhail’s suggestion and delete the “fixed totality” part? You’d get something like,
“If human beings were to be removed from the universe, there would still be the same non-human objects.”
With Mikhail, I think that something like this thesis will still do the work that Braver puts R1 to. Interestingly, it starts to shade into Harman’s R7.
Does that work?
I think the point of the objects being “fixed” is to help establish anchor-points for questions of reference (in a metaphysics that cares very little for questions of Occasionalism). As the reference is Putnam – and I could be wrong – I have very little faith that the idea that the planet you and I call “the morning star” (or the “evening star”) is a different object time t1 and t2 is something of great interest. The point of this kind of realism is that objects rigidly are preserved as designated.
Perhaps though someone could fill this out for me.
Sure, I suppose that something like
is as good a place as any to start. I’ll accept that. For what it’s worth though (and I mean this in the friendliest way possible) such a starting point strikes me as exceedingly abstract; it’s the kind of counterfactual that people — following Hegel, actually — usually criticize Hobbes and Rawls for using; i.e. the a priori that allows one to to construct this possible world turn out to have be the hard won’ a posteriori of our contemporary position. So it may in fact be question begging. Again, I’m not trying to be confrontational, I’m just not sure how much sense we can actually make of this suggestion — other than as a counterfactual kind of argument for the status quo (a refutation of exotic Meinongian universes where teacups, quarks, unicorns, and square circles all have the same ontological status, maybe). Other than this critical function, though, I’m not sure wht productive traction this thesis really affords.
What do you think? Is there there something specifically productive about claiming a counterfactual world of natural kinds?
Levi, I have some questions concerning your terminology. I am fascinated by this “tool-being” program, but I am confused on some central issues.
Why do you put squarequotes around “translate” if you feel you are not anthropomorphizing? What does it mean for a rock to “receive” an object? Not to sound rude, but how is this not metaphorical re- description of scientific concepts of causality? If philosophy is meant to help us understand the world better, I just don’t see how talking about the translation of differences will help us better understand how the plant interacts with other objects independently of human interaction.
You say a difference “acts” on another object, but does this not play on our human understanding of “act”? If you are trying to really do object-oriented philosophy, it seems like the best route would be to go with the vocabulary of causal-nomological theory, devoid of references to human intentionality. We can describe the plant’s interaction with the world down to the sub-molecular level in terms of complex causal chains, and this in turn gives us a better understanding of how objects are in themselves, but what does talk of “translation” do for furthering this understanding?
I feel like the vocabulary of science already gives us the linguistic toolkit necessary to describe the world in non-human terms, what else is left? Talking about objects as if they have a monad-esque “point-of-view” from which they “act on” other objects and actively “translate” differences, seems hopefully anthropocentric compared to the nitty-gritty technical objectification science employs on a daily basis.
I am sympathetic to the overcoming of crude scientism, and perhaps I am confused on the central motivation to discuss object-object relations in such metaphorical terms, but I can’t shake my feeling that such talk is simply that, metaphorical.
Orestes,
I think we would be hard put to find an example of science or philosophy that isn’t riddled with metaphor, so I’m unphased by this sort of criticism. In biology, for example, we talk of organisms adapting to their environment. This implies intentionality. Yet there is no intentional in processes of natural selection and random mutation. Likewise, in physics we talk about bodies being attracted to one another, which implies a reference to human emotions, yet at the physical level gravity has nothing to do with attraction. These metaphors help to draw our attention to particular phenomena and grasp it.
In my ontology the term “translation” is adopted to draw attention to the role that multiple differences play in the genesis of phenomena. The “cash-value” of this principle only becomes apparent in relation to what I call the Hegemonic Fallacy. The Hegemonic Fallacy consists in the reduction of difference to one difference that makes all the difference. Maybe I can clarify this point clearly in terms of Kant.
What is Kant’s remarkable philosophical achievement? Using the language of my ontology, Kant’s great achievement was to be among the first to recognize the role that translation plays in the production of knowledge. For Kant, our minds are not simply passive pieces of wax that receive the imprint of data that affects the mind, but rather the mind actively translates the data that affects it, producing something new or in addition to the data itself. Kant’s name for this process of translation is “synthesis”.
By recognizing the role played by translation in cognition, Kant was able to denounce the Hegemonic Fallacies and violations of the Ontic Principle (the thesis that there is no difference that does not make a difference) made by philosophers of the rationalist tradition and the empiricist tradition. The thinkers of the rationalist tradition committed the Hegemonic Fallacy and violated the Ontic Principle through their belief that we can reason to truths about the nature of God, the world as a totality, and the soul through pure concepts. In other words, all difference, for them, came to be located in pure concepts. In doing this, they ignored the manner in which concepts are translated by intuition (i.e., that concepts must be synthesized with intuitions to produce knowledge). Similarly, the empiricists violated the Ontic Principle and committed the Hegemonic Fallacy by locating all difference in intuitions or impressions, ignoring how intuition is translated by concepts. Kant differs from these traditions by virtue of his great attentiveness to the multiple differences involved in the production of phenomena for mind (differences pertaining to concepts, intuition, reason, etc).
My thesis differs from Kant in that where he restricts translation to mind-world relations, I hold that translation takes place among all relations. I believe that the gentle reminder that all relations involve translation helps draw our attention to the role that multiple types of differences play in relations. In my view, differences are often ignored and everything comes to be reduced to one type of difference. Take the example of Lacan. Lacan, in Seminar 20, says “the universe is the flower of rhetoric”. Lacan’s thesis is that being is a product of language. In advancing this thesis he has reduced being to one type of difference. But what he says is simply not true. While language plays a deeply important role in how we carve up the world– and is therefore not a difference that we should ignore –there is also the objects of the world themselves, our interactions with one another, our brains and biology, etc., etc., etc. Similarly, Lacan argues that the subject is a product of language. Again, language is important to our status as subjects, but so is our neurology, our biology, our interactions with a physical relation to the environment, and so on.
In my view, theory often suffers from confusing its abstractions with being itself. We begin by first isolating some set of differences to investigate, and then forget that we excluded a whole host of other differences. The concept of translation, I believe, is a gentle reminder not to forget these other differences or multiple determination.
Why, then, do I use words like “translation”, “difference”, “act”, and so on? Philosophy has a history and is also meta-theory. As meta-theory it’s important to choose terms that are broad enough to cover a wide range of phenomena. If I do not use the term “causality”, then this is because not all types of translation are causal interactions. I thus need a term broad enough to capture both causal interactions and other types of translations such as the one I described above in relation to Kant. Moreover, talk of causality often suffers from forgetting the multiple determinations involved in a causal interaction between two objects. In order for one object to be capable of producing an effect in another object, that second object must have certain affects. By “affect” I am not referring to “emotion”, but rather, following Spinoza and Deleuze, I am referring the capacity to act and be acted upon. A neutrino, for example, is unable to affect many other objects because of the nature of its constitution. Simply put, the affects an object has translate the causal impact of another object and render it possible. This, I think, is often ignored in many discussions of causality.
One final point. I am surprised, given my prior post, to see references to science and what it can and cannot tell us. The realism of Object-Oriented Ontology or what I call “Onticology” is not a scientism, nor does it place the real on the side of science and nature, and the unreal on the side of anything human. Humans and human phenomena belong to the real every bit as much as a quark on the side of the universe. The thesis of Object-Oriented philosophy is simply that humans aren’t included in all relations.
I know this might be an example of my infamous “nitpicking” but you say:
I am really puzzled by this explanation. It seems that according to you because we find metaphors everywhere, it is absolutely okay then to use metaphors in whatever way possible, yet as Orestes pointed out, and I think that was his main point, you still hesitate and put “translate” and other terms in quotation marks? I mean, sure metaphors are everywhere, as you say, but that does not necessarily excuse sloppy philosophical terminology, does it? If we were to follow this logic, we can say: “Well, language is often very ambiguous. When I say ‘physical attraction’ I don’t know what it means without much context. Therefore, I shouldn’t worry about ambiguity in my formulations.” It’s a sort of “he who is without sin throw the first stone” argument, isn’t it?
Now, as I pointed out many times, I see where you are coming from with this “I argue that X” kind of style, but it’s hardly going to make much impact when there is a number of others saying “But I argue that Y” which is the case in our discussion of Braver’s theses on realism. Essentially, he presents the various discussions of realism and cites various sources as a starting point of his own investigation and you counter-cite yourself and your propositions and then take Braver to task that he does not consider your kind of realism which is hardly fair, is it? As I suggested, if your understanding of realism does not fit with any of the presented features of what the majority of those who use the term “realism” mean by it, then it’s very likely that your position is not realism – this is of course an argument from authority (“this is what scholars say realism is, and yours isn’t on the list”), but it isn’t very different from your argument which is also an argument form authority (“My position is X and it is a type of realism because I wrote many posts claiming that it is”)…
As I read my comment again, I realized it might come off as too harsh and dismissive if you don’t know the context of Realism Wars and such, so just a disclaimer then – we’ve discussed these issues over many months/posts and it’s in the good spirit of philosophical debate.
Mikhail,
My point about metaphor is not that we should applaud the sloppy use of language, but that philosophy often makes metaphorical use of language in stipulating concepts. Take Aristotle’s term “category”. In Greek this term actually has legal origins. Aristotle puts it to very different use. As philosophers and theoreticians, we constantly stipulate the meaning of the terms we use, taking words that have different connotations in ordinary language and putting them to specific theoretical uses. The reason I put the term “translation” in scare quotes was to indicate that I’m using the term in unusual technical fashion and that the reader should take care to pay attention to precisely how I am using the term.
As for debates over what realism is, two points are in order. First, as someone else noted in this thread, Braver’s account of realism is contentious as he primarily cites anti-realist descriptions of realism to define realism. That aside, it seems to me that there are not six features common to variants of realism, but two. The common feature of all realism is the mind-independence of objects and the thesis that there are relations among objects that do not involve the human in any way. That’s it. Apart from that, there are a wide variety of realisms that diverge markedly from the sort of realism that Braver describes. Whitehead, for example, is a realist but doesn’t resemble Braver’s account of realism in any way. DeLanda is a realist, but doesn’t resemble Braver’s account of realism in any way. Spinoza is a realist but doesn’t resemble much of what Braver outlines.
There is a sort of inequity in Braver’s book. It is recognized that there are a wide variety of different anti-realist positions and debates among anti-realists themselves, and he approaches these anti-realists on a philosopher by philosopher basis. From the realist side of the spectrum, however, realism is reduced to a caricature and we don’t get much in the way of individual philosophical voices articulating different varieties of realism. In stating this, I am not rejecting the value of Braver’s book. Braver set out to write a book on Continental anti-realisms, not Continental realisms. It is thus fair for him to vaguely gesture at realism and be rather reductive as to what he’s referring to. However, I don’t think it’s very fair for you to say “well perhaps you’re just not a realist” when I mark my philosophical difference from the version of realism Braver describes. There is a double standard here. The Continental anti-realist like Foucault gets to say “I’m an anti-realist, but not of the Kantian sort”, whereas suddenly the realist has to be pigeon holed in the category of a rather vulgar Russellian version of realism without being able to argue for a very different version of realism.
I mean, c’mon, we could make the claim that Heidegger’s use of the term “dasein” is metaphorical in the sense that in German the term refers to existence, not to the human. Heidegger chose the term “dasein” both because of its connotations of existence, but also because he was attempting to depart from the prior philosophical tradition and was looking for a term to designate human being-in-the-world without all the subjectivist and humanist connotations of the word “human” in the prior philosophical tradition. He carefully stipulated the conceptual content of the term “dasein” within the scope of his philosophy. I do something similar with the term “translation”. On the one hand, I choose the term “translation” precisely because it’s connotations of an ordered transformation of something from one medium to another medium as in the case of translating a book from French to English. Such an ordered transformation produces new resonances of meaning and sense when placed in the new language, as well as a loss of information and meaning or differences from the original meaning of language. I then give the term a technical meaning, broadening its scope beyond language to mark the way in which interactions between differing meanings produced ordered transformations.
Does that mean that Kant is a realist then? As someone pointed out, Kant does affirm that existence of mind-independent objects (things-in-themselves) that affect us and give us mind-dependent objects. I’m not sure if he denies relations among mind-independent objects, it doesn’t seem that he would from what I remember. In other words, is your definition of realism then too minimal and thus too inclusive? Who would be an example of anti-realism then? Berkeley?
It’s not an on-off thing. In Braver’s account, Kant is a realist in respect to some of the theses and an anti-realist in respect to others. The really amazing thing is how he’s able to show everything that follows (well: Hegel, Nietzsche, the early and later Heidegger, Foucault, and Derrida) are the working out of problems with Kant’s combination. That’s where the theses really get friction in Braver’s account.
Kant is a realist in his affirmation of things-in-themselves, however, it is a very minimal sort of realism. Beyond their mere existence and the fact that they affect the mind, Kant prohibits knowledge of much of anything else about them. I don’t know that Kant prohibits relations among things-in-themselves, but from a Kantian perspective we can’t know anything about these relations because all the important relations (in the categories) are belong to mind in such a way that we cannot know whether or not these types of relations also belong to the things themselves. I think there’s a strong and a weak version of Kantianism, both of which have ample textual evidence. The strong version of Kantianism argues that things-in-themselves do not in anyway possess characteristics like time and space or the categories. The weak version would say “things-in-themselves might be structured in the same way as phenomena, but we can’t know one way or another and, for all intents and purposes, it doesn’t matter pragmatically.”
You might be right that my definition is too minimal. It was actually me that pointed out that Kant affirms the existence of mind-independent objects that affect us and give us mind-dependent objects. My objection to Kant is not his talk of mind-object relations, but his restriction of philosophy to mind-object relations, excluding any talk of object-object relations that do not involve the human.
Oops, so I was quoting you to make a point to you, my bad – with so many comments it’s hard to keep track sometimes. After this reading group is done, maybe someone can collect it all into one large PDF or something?
From what I know of Kant, I think you are right about Kant’s restriction, but he still fits with the minimal definition of realism then, that was my point, that is, if the definition is so wide than it might not be very useful and all sorts of folks could be realists.
I’m not sure how persuaded I am about “partial realism argument” (if I can call it that) that is, I can be realist about one set of things and an anti-realist about another set of things, especially if we accept the wide/minimal definition that all realisms only have two theses to qualify (may I call them LS1 Independence and LS2 Object-Object Relation?). It seems that if we accept LS1-2, then we accept a certain ontology/metaphysics, a certain”world,” if you will, a certain totality of objects and their relations, then it is impossible to be anti-realist about some set of things in this type of world, because to be anti-realist would mean to deny LS1-2, i.e. in this scenario, it seems, you’re either realist all the way (LS1-2) or anti-realist all the way (ALS1-2).
Levi, I still think that throwing the word “metaphor” around is a bit of an excuse and makes things confusing – to say, for example, that a term might have several meanings is not to say that one is “literal” and another is “metaphorical” as if there is only this one pair of options available. But I think this angle of approach would take us away from the discussion of the book here, so I’m willing to drop this objection and let Orestes deal with it since he brought it up.
I do think that the issue here is that of self-identification vs. general description, i.e. let’s imagine a simple case in which I state that I am a realist and I define realism as X and you state that you are a realist and you define realism as Y. In one scenario, we are both realists because we can see the way to combine X and Y (realism then is X+Y). In another scenario, neither one of us is a realist, or only one of us is a realist while the other is mistaken in his self-identification. But it seems to me that in all cases, we must have an operative definition of realism to be able to make all of these judgments – the question is then – where does this definition come from?
We can have Lee answer for himself in this case, but to say, for example, that a particular presentation of realism is a caricature, one needs to know what real realism looks like (to compare and to conclude that something is a caricature) – if this real definition of realism comes from a number of postulates, then where do these postulates come from and, more importantly, what are the criteria according to which we may compare Braver’s postulates and Levi’s postulates and decide which ones are closer to real realism? I hope this makes sense.
I’m not really seeing the problem with Levi “stipulating” operational/theoretical uses of words. In a totally different context, Ernst Nagel wrote something to the effect of, “well, the opposition between realist and instrumentalist views of theories is really a conflict over which mode of speech one prefers.” Or something like that. Really, I completely understand Mikhail’s point regarding standards/criteria regarding a fictional or operational “real realism,” but why must the onus be placed on Levi’s shoulders (other than he claims that his realism doesn’t match up with R1-R6, and really, my response is a friendly (a) who cares? or more to the point (b) when you use the term realism just what is it you mean? Fair game, in my book.
On the other hand, I think Mikhail is right to try to account for a “common denominator” or core position. Realism–rather broadly speaking–seems to me to entail the the view that our best theories are either true or *approximately* true representations of the nature of the world. Realism then, would also be committed to the view that progress in say, science or philosophy or whatnot is progress towards having true theories. So, naturally, a correspondence theory of truth emerges here also. But certainly there are other starting points as well…
One of the things that comes up with Kant (and in the discussion above or somewhere) revolves around inferences, viz., the truth of say, realist theories is the best available explanation for their explanatory power (and I suppose something about predicative success). Yet, our best explanations may not be true ones. A good explanation certainly entails other stuff besides truth, no? Moreover, we could all come up with cases in which truth isn’t even necessary in a good explanation. So, here’s the thing, if our most workable or “best” explanation is not by necessity true this creates some problems, perhaps, but the realist could wait around for necessary conditions for truth, or course. Ok, here’s my point, then. What we see with Kant and later on, with a bunch of different anti-realists, is the addition or subtraction of a variety of views; be it coherence theories of truth, metaphysical pluralism, or epistemological views that straight jacket/deflate any sort of um… speculation about metaphysical “stuff.” For their part, perhaps “realists” add say, strong theories of truth as correspondence to reality, and flat ontologies or what have you. The real question then would seem to be centered around the possibility of identifying a core approximation of “realism” so that the fist-pounding of the “Realism Wars” isn’t repeated.
Just a thought.
That’s why I think that Braver’s move is okay with me since it is only the beginning of the conversation, if his first chapter was in fact his last chapter, then I think we would have a case. We have to start somewhere – true, Levi’s point of reference might be as good as any, but then again I wonder if the point of departure really is so arbitrary? I mean there must be some, even if very minimal, designation – I understand that Levi wants to retain the label “realism” for his philosophical position and therefore the minimal definition works for him, but Independence Thesis and Object-Object Relations that are the main elements of the minimal definition are still in need of justification and cannot simply be dogmatically asserted (“dogmatically” in Kantian sense, i.e. without deduction). When I ask a simple question like “Is my hatred of country music real or not?” I don’t mean whether it is mind-independent or any such thing, I mean “is it real or imaginary?” or “is it real or am I faking it?” If we begin with the simple meaning of terms, Levi’s “minimal definition” is already not so minimal, but in fact is very much loaded (one might argue that it is loaded in his favor), as well as Braver’s point of departure. So my point is this then: yes, both Braver’s R1-R6 and Levi’s minimal definition might be conceived as starting from an arbitrary point (which I would contest), but what sort of “loading” is taking place when Braver or Levi begin where they begin?
Another metaphor: what is won when the hill is scaled, the enemy is vanquished and one’s flag is planted in “realism?”
LS: For Kant, our minds are not simply passive pieces of wax that receive the imprint of data that affects the mind, but rather the mind actively translates the data that affects it, producing something new or in addition to the data itself. Kant’s name for this process of translation is “synthesis”.
Kvond: but it was never the case that the wax metaphor was one of mere passivity. Always the quality of the soul (wax) was found in the very degree of its capacity to respond to the impression, and this response was always included in the image:
Vostra apprensiva da esser verace
Your perception from some real thing
tragge intenzione, e dentro a voi la spiega,
an impression takes, and in you it unfolds,
sì che l’animo ad essa volger face;
so thus the soul to turn it brings;
e se, rivolto, inver di lei si piega,
And if, so turned, toward it she molds,
quel piegare è amor, quell’è natura
that molding love, that nature ‘tis
che per piacer di novo in voi si lega.
which by pleasure fresh in you it binds.
Poi, come ‘l foco movesi in altura
Then how fire upward glides
per la sua forma ch’è nata a salire
by its form being born to climb
là dove più in sua matera dura,
t’where most within its matter it abides,
cosi l’animo preso entra in disire,
thus the soul so pressed enters in desire,
ch’è moto spiritale, e mai non posa
a spirit motion, that wilt not rest
fin che la cosa amata il fa gioire.
‘til the thing beloved makes it ‘joice.
Or ti puote apparer quant’è nascosa
Now should appear to thee how clouded
la veritate alla gente ch’avvera
be the truth with men who deem
ciascun amore in sè laudabil cosa,
every love itself a lauded thing,
però che forse appar la sua matera
perhaps because its matter seems
sempre esser buona; ma non ciascun segno
always to be good, but yet not every stamp
è buono, ancor che buona sia la cera.’
is good, even if be good the wax.’
Canto XVIII, lines 22-39 Purgatorio [hopefully the fonts and formating of the verses works]
What Kant did was unhinge the waxing movement of the soul towards the very coherence that was producing it, that is, he privatized the coherence unto itself, thereby cutting it off from organizational processes that were occurring outside and beyond it. If there is a claim for realism, it is the claim not only that science describes things as they are (really an almost inconsequential point that really becomes a question of authority), but that our potential for agreements lies not only within the specific horizon of our variously defined (and historically contingent) normativities, that there are real processes of coherence and organization that add to the value of human life, that open it up to kinds of incorporation iand external synthesis.
follow up: [okay, only some of the formating carried through] I hope the Dante quote points out that the perception of an impressive “some real thing” is really an event for a discernment, one which directs us across a realism of valuations which of their very nature participate in the real objects we direct ourselves toward. Dante’s Realism is really a realism of valuations as necessarily extra-human ontological combinations.
Thanks for your input, Kevin. I erased the pinbacks to your post on Dante (and some of the pinbacks that WordPress keeps created on constant self-reference of our own posts etc), I don’t know if you noticed, to keep the “Recent Comments” column a bit cleaner.
I had not noticed how ping-backs work on other sites. Delete all you want. I self-reference and cross index with the knowledge that people might be reading a post a year from now and have very little knowledge of the context of the post. But I have little understanding how WordPress handles these things, and I have to delete sidebar pings from my own blog as well.
Wow, lots of comments, very interesting discussion so far even if over my head in spots (and what were the Realism Wars you refer to?). I’d like to add a few things if I may.
Rorty used to say that his pragmatism overcame the “stale” realism-antirealism debate. I always liked the sound of that — the various antirealisms always struck me as unhelpful recoils (in the McDowellian sense) from realism — but Rorty himself eventually disappointed me. When he insists that the world is “not found but made” (etc.) he strikes me simply as falling onto the antirealist side of the very dualism he promised to overcome. If we must have a slogan, Putnam’s “the mind and the world jointly make up the mind and the world” seems as good as any.
So I’m not so concerned with nailing down “realism” as R1–R6, nor with replacing them with A1–A6, whatever they may be (I don’t have the book). Ultimately, or even proximately, “realism” so construed is incoherent (or better: indeterminate, at this stage, between incoherent, false, and uninterestingly true). Again, Putnam seems to me to draw the best caricature (in the good sense). That at least is how realists speak when they disrupt our conversations, demanding (with Kripke) to know what Pierre believes about London, or (with Stroud) how we know that the sky is really blue when maybe it isn’t.
This caricature is thus all we need for Braver’s task, which as I understand it is to give a history of how continentals have tried to get realists to stop doing that. Naturally this will involve flirtation, as I myself have flirted, with various forms of “antirealism.” And indeed, who cares where the terminological chips fall if we end up with something we like. “Pragmatic” realism [Putnam] as opposed to “Cartesian” realism [Margolis, in opposition]? Or maybe “objective” “idealism” as opposed to “subjective” idealism? Whatever. As long as the key dualisms are overcome, I’m happy.
My own view, just for reference you understand, is an ungodly amalgam of post-Rortyan pragmatism, quasi-Nietzschean perspectivism, and a pseudo-hermeneuticized late-Davidsonian triangulatory semantics, garnished with gnomic appeals to Wittgenstein every five minutes. So my account of belief makes essential reference to “objective truth,” and of course there are truths we could never know; but o.t.o.h. whenever I hear the term “mind-independent” I reach for my SuperSoaker.
In general I think of realism and antirealism as attitudes more than doctrines. We all have Platonism in our distant (or not so distant) pasts. How should we react when everything solid starts melting into air? Realists panic, seeing rationality and indeed civilization going down the drain. Antirealists want to help the process along, and get rid of those fixed hierarchies once and for all. So when realists grasp a lump of reality they wish to preserve as ideally real, antirealists reach for the potato masher, hoping in the end for a smoothly continuous substance. Up to a point, the antirealist is right: no ideally real lumps! But up to a point the realist was right too. The Platonic picture didn’t come out of nowhere, but as a reaction to sophistry; and it would be a mistake to think of everyday locutions concerning objective reality to require pre-emptive philosophical qualification (itself a lesson which cuts both ways). Ultimately we will need to make use of both attitudes, when properly constrained. Keep your potato masher around, but don’t feel you need to use it whenever you see a lump forming. Maybe, without realists jumping in to preserve it, it will go away on its own, as others form and dissipate in turn. (Maybe this is what Levi is saying, but I’m not sure.)
Maybe we will get to this, but I think a key issue is that of the approach of each philosophical attitude toward self-reference. Antirealists (esp. relativists and the relevant sort of skeptic) are indeed, as realists love to point out, caught in a conundrum when asked how their views apply to themselves. Or at least they must do some work in order to answer a question which realists take themselves to be able to answer perfectly straightforwardly (Q: Is realism itself really true? A: Yes.). On my view this realist answer isn’t so straightforward, ultimately requiring a Platonic regress-stopper, and antirealists can often do better than saying things like “yes, “truth is what works” is true, because it works.” As I said before, whoever can occupy that middle ground can call their view whatever they want.
I won’t go into how my view is superior to all others, but I see my time is up and I haven’t appealed to Wittgenstein yet. Wittgenstein pointedly denied promulgating philosophical doctrines in favor of showing us different pictures, in the hope of getting us to “see connexions” and thus be able to “find our way around.” Yet each picture is itself partly constituted by doctrines, which in that context can be argued for if it seems appropriate to do so. What makes this view hard to approach is (unfortunately) that until the whole (over which, as LW says, “light dawns slowly”) is grasped, its advocate looks like he’s talking out of all five sides of his mouth, arguing for some things at one point and abandoning them later, only to go back to them further on. It can be quite confusing. You have been warned 😎
Dave: “Rorty used to say that his pragmatism overcame the “stale” realism-antirealism debate. I always liked the sound of that — the various antirealisms always struck me as unhelpful recoils (in the McDowellian sense) from realism — but Rorty himself eventually disappointed me.”
Kvond: Rorty and Davidson’s Anti-Realism came eventually better referred to as “Non-Realsm” (I believe). I’m sure that you know this, but it bears repeating: the reason for this is that the HUGE question that characterized Anti-Realism from the Analytical side (and I have pointed this out before), is that of justification. The Anti-Realists became Non-Realists because they refused the APPEAL to the Real (some real properties, facts) that determine the truth of sentences that necessarily must ALSO have a sociology, historically contingent aspect. This was the beauty of Rorty’s Mirror of Nature analogy I think. Once though the question of the justification of the truth of sentences is let go of, then so many of the Realist intuitions about the firmness of the world can be reclaimed. In this sense the Anti-Realist strain that comes from Rorty/Davidson is really merely an epistemic restraint on appeals of justification itself. This fits nicely within Larval Subject’s Latourian principle of Translation, I would think.
Turning Anti-Realists into Non-Realists may relieve some of the problem that seem to be implicit in the firmness of their own claims, turning instead to a certain functionality of rationality (as Davidson appeals to), something that is an evolved property/capacity.
Well, again, I like what Rorty *promises* to do, but at the level of what exactly to say w/r/t Davidson and truth, I am sympathetic to McDowell’s verdict that Rorty’s pragmatism fails by its own lights (as in his article in Rorty and his Critics, and the way I have put it here is that he is too keen to avoid even the appearance of backsliding into realism (as he accuses McDowell of doing). Some of this has to do with justification, about which very few people agree with me. Let’s see if or how it comes up later; I am reluctant to get into too much detail in advance.
Alexei: “Is there there something specifically productive about claiming a counterfactual world of natural kinds?”
Kvond: Is this not the essential act of scientific ceteris paribus? That is, the assumption that conditions are the same with or without us grounds all our other, more specific, ceteris paribus descriptions, does it not?
I don’t think I follow, Kevin. Could you elaborate a little more? Maybe this is due to the fact that I have a rather narrow understanding of the use of ceteris paribus clauses, but I would have thought that they simply signal variability of certain elements in, or parameters of a law. For example, water boils, ceteris paribus, at 100 degrees Celsius. What’s variable, of course, is elevation — only at sea level will water boils at precisely 100 degrees. So the proper nomological form for a law concerning boiling water would have to introduce a variable for elevation/atmospheric pressure.
Does that make sense? Or have I missed your point entirely?
Alexei,
Of course it is more narrow. What I suggested is that it functions narrowly due to a much broader conception, that is, the conditions that we are measuring, talking about, interacting with would be the same if we were not doing so (this assumption is abrogated if one discovers that our interactions themselves are creating problems with the observation or conclusions). You ask what good does such a wide-scale counterfactual assumption make, and I offer that it may ground all of our counterfactual observations and thus claims for objectivity itself.
Not only does water boil under such and such a condition, apart from other variables, but it does so apart from whether human beings exist. This seems like an essential assumption, which I would group under a causal appreciation of how persisting features of the world affect both you and I when we discuss a subject matter objectively. Does that make sense?
Ah, I see now. Fair point. I concede it.
Nicely said. If we all could be capable of such sentences as a regular (potential and actual) part of our discourse.
-Sorry, but I can’t post a lot given my lack of internet and computer here in Maine. (also, leaving for Italy early July- I can only comment until then, but I will eagerly await to reply to late Foucault and Derrida commentaries)
-Wow! Thanks for the additional info Dr. Cogburn. Always excellent.
-One question: Why doesn’t Braver include his description of Anti-Realism into this first chapter? If Realism and anti-Realism are suppose to be so complementary (in the negative sense), then why completely ignore the critiques of anti-realism. I assume that Braver attempts to answer this historically, however, I am skeptical that an analysis of late-Wittgenstein, Quine, and Dummett would not be useful right here at the beginning.
Pingback: More on the Antigone Complex « Frames /sing
This is a really great opportunity to participate in this reading group. Hopefully this will give rise to many, many more such endeavors. I feel like I am participating in an upper-level course – for free!
My question is much simpler (perhaps by necessity) than all of the others, but please humor me:
Let’s say I am a realist and my colleague down the hall is an anti-realist. Furthermore, we often have occasion to debate the merits of our positions while making epistemic claims that may/may not be justified. At the end of the day, though, the both of us will go out into the same something and navigate round this something in pretty much the same sorts of ways. So, if one isn’t privy to knowing which metaphysical theory my colleague and I support, how can it be discovered without asking?
How can it be discovered without asking? It cannot. I think you are conflating “holding a metaphysical theory” and “metaphysical theory” in your question. If I am holding a metaphysical theory that onions contain a special evil ooze that is slowly destroying my brain, you cannot see it in my practice of ordering every food without onions, you might just think that I don’t like how they taste or I am allergic etc etc. The question here is not how my metaphysical views affect my everyday life, but what my metaphysical theory is – in order words, to evaluate my theory in terms of its “noticeability” is not quite fair, I think. On the other hand, if I take your question to be a sort of general “Why does it matter if I am a realist or an anti-realist?” I think it’s a great approach and I would think that most people would be able to answer it in terms of “It does because X” and hopefully we can eventually get to it.
Mikhail (or anyone else),
Can you expand your take on the diference between “holding a metaphysical theory” and “metaphysical theory”?
I guess what I meant was something like this: “realism” is a metaphysical theory, “being a realist” is holding a metaphysical theory (to be true). I suppose you can also distinguish it from “claiming to be a realist” as a claiming to hold a specific metaphysical theory like realism. Clearly, going back to your question, being a realist is not going to affect the way you ride the bus, but it might affect the way you argue for or against a specific moral position, or justify a certain set of beliefs, or do science etc etc. At the same time, this “holding a metaphysical theory” does not really influence that theory itself – if a certain realist proposes things that are far from realism, it’s not realism’s problem, is it? Just for example…