Since the discussion about the possibility of error is much-discussed at Uncommon Descent, I thought it might be interesting to see how Josiah Royce develops his argument concerning “the possibility of error” in his The Religious Aspect of Philosophy (1885). (I’m using The Philosophy of Josiah Royce, which I found recently in a used-book store. I assume that no one here is too concerned about quotations or citations, but those are available on request.)
Royce’s question here is, “how is error possible?” — and by ‘how’ he means, “what are the logical conditions for the possibility of error?” An error, he points out, is our recognition of the failure of a judgment to agree with its object. How is possible for us recognize that our judgments have failed to agree with their purported objects? The puzzle goes as follows: on the one hand, if the object is entirely within our cognitive grasp, our assertion about it would fully correspond to the object — in which case, there would be no error. On the other hand, if the object were entirely beyond our cognitive grasp, we would be unable to recognize the lack of correspondence between the judgment and the object — in which case the error would be unrecognizable. So our ability to recognize errors as errors requires that we have “partial knowledge” of the object. So what is partial knowledge, and how is it possible?
[It will not surprise anyone here who knows how I think to learn that, from my point of view, the above is more-or-less sound, whereas the next bit utterly goes off the rails.]
What is required, Royce thinks, is that both the judgment and the object are contained within some larger, more inclusive thought that can compare them against them against one another and notice the correspondence (or lack thereof) between them. And since there are infinitely many errors, the inclusive thought must be all-inclusive — it must contain all possible judgments and their objects. And that in turn must be the Absolute Knowledge and Absolute Mind of God. (Didn’t see that one coming, eh?)
TL;DR version: there are errors, therefore God.
That sounds like the creationism and ID arguments.
It is very hard to say what that means.
We make judgments of truth. But we also make pragmatic judgments, moral judgments, artistic judgments — and probably many others.
There seems to be a tendency for people to think that all judgments can be subsumed under “judgments of truth.” And I suspect that may be what leads people to a belief that there must be absolutes.
I suspect you’re right about a certain temptation to which absolutists are prone. But Royce’s argument doesn’t depend on that temptation, since it could be easily restated to make explicit the restricted focus to judgments of truth, regardless of the other kinds of judgments we make. In other words, so long as one admits that we make any judgments of truth at all, Royce’s argument retains whatever validity it has.
Of course, in recognizing our error we can still be in error about what our error was. It’s not really partial knowledge that informs us of our error, it’s experience subsequent to that error.
llanitedave,
True, but Royce’s point is that there has to be a ‘partial knowledge’ of the object in order to recognize that there was error — even though that partial knowledge is not sufficient to determine what the error was. It’s that ‘partial knowledge’ that Royce is trying to explain. Even if we all agree that his argument for the Absolute Mind of God isn’t convincing, it’s still worth taking some time to notice that it’s an interesting question he’s gotten hold of.
Later on his life, under a more thorough-going influence from Peirce, Royce’s emphasis on the Absolute Mind of God is replaced by an emphasis on the community. I find that a significant step in the right direction, though not completely satisfactory.
Anyway, given how frequently the “existence of error” is bandied about at Uncommon Descent, I thought it might be interesting to see what Royce originally meant by that term, and how he used it.
From what I can see, Royce didn’t have much exposure to the basic concepts of probability and statistics that were still being developed but were already fairly well understood, even during his time.
Many further developments came later in the 20th century; but even today high school students can get a pretty fundamental understanding of probability distributions, sampling distributions, and inferential statistics, including Bayesian statistics, in an Advanced Placement statistics course. These courses are accessible to bright students even in the 9th grade.
But trying to make inferences about things for which one can obtain no data doesn’t get one very far along a spectrum from ignorance to certainty.
Writing as the chief opposing protagonist in the endless error debate on UD – thanks for this. Where does KF get all this stuff about “error exists” being a self-evident truth – or is it his own invention?
There seems to be something implicit in this that you only have one shot. You are either right or wrong, and that’s that. But information flows continuously, and you don’t forget your memory of your initial perception. I thought it was a tiger, but turns out I was wrong. Looking more closely, I see it’s just an arrangement of shadows. Am I sure it’s not a tiger? Yes – see for yourself. We are both agreed. Let’s get everyone in. Not A Tiger. Agreed, everyone? [Roar, roar, chomp, chomp] Oh, bollocks, we were all wrong!
But does the actuality of the tiger require an observer at all, someone who can be right or wrong about it? I don’t see why, though Berkeley might say it does. To me, the possibility of error requires only that something exists that can be wrong, and does not require an observer who is right.
I don’t know what logical conditions suffice for error, but practically, there has to be a perception to allow the possibility of an incorrect one. The object must be capable of perception, our perceptive faculties must be capable of deception, and we must access alternative perceptive faculties (our own or others’) to permit detection and correction of the error. One has to discard the erroneous position (and recognise it needed discarding) in order to replace it.
Someone at UD said something to the effect, “Error only exists in the mind.” Outside of cognitive human beings, where does “error” exist? Would we think that any other animal has the concept of committing an “error”? Material objects just do what they do. There is no “error” in the ongoing interactions of materials in the universe.
IMO, under materialism, the idea of truth and error, of perfection and imperfection are incoherent. There is only “what is”, not “what should be”. There is only “what I think” not “what I should have thought”. There is no “what the universe should have done”, even when it comes to generating thought.
William J. Murray,
I’d say a bird that flew into a window had made a mistake. I’d expect the bird would agree, but clearly we cannot compare notes. God could provide the definitive answer.
“If a tree falls in a forest…”
I’ll take that as an admission by WJM, of his own relativism.
I think this ‘partial knowledge’ of the subject is where he begins to get off track. Taking the word ‘error’ not in an abstract philosophical sense, but the way it is actually used, it merely implicates any judgement or opinion that diverges from our present judgement or opinion. There is no requirement of ‘knowledge’ (in the contrived philosophical true-belief sense), whether partial or complete.
When I read that talk of “the logical conditions for the possibility of error”, I pretty much assumed that he has to be looking at error in the sense of a false proposition. But then comes that talk of “the failure of a judgment to agree with its object”, and I wonder what that can mean. If Royce is concerned with logical conditions for error, then presumably “its object” must be a reference to a proposition.
That whole way of talking about error and judgment seems to presuppose the existence (in an ideal or platonic sense) of a system of propositions which describe reality in every detail. So it seems to me that Royce is implicitly assuming the kind of absolutist view that he claims to show.
To me, that way of looking at things is absurd. But perhaps that’s why I am not a philosopher (or, at least, not a professional philosopher).
An example of error
Earlier, I considered posting this example to the music thread. But it fits even better here.
In the IBM mainframe world, there’s a utility program with the name “IEFBR14“. Its purpose is to do nothing.
A little background. Mainframe systems use JCL (Job Control Language) to define jobs (programs to run). The JCL interpreter does some useful related things, such as allocating disk space. The purpose of the IEFBR14 program was to allow use of those JCL interpreter functions, when you don’t actually have a program to run.
The original version of IEFBR14 consisted of a single instruction. In Assembler, it was:
BR 14This is the equivalent of a trivial higher level language program containing the single instruction:
As you can see, this is simplicity itself. It is the simplest possible, and therefore the most perfect program that does nothing. One might almost suggest that as a self-evident truth, and it probably seemed that way to the programmers who wrote that simple program. Unquestionably, such a program would have been considered error free.
A few years later, there was a bug report for the program. IBM responded with a bug fix. The corrected program was now:
SR 0,0 BR 14The equivalent, if written in a higher level language, would be:
return (0);What had happened in the interim, was that the computing milieu had changed, such that it was now important for programs to give a return code (or exit code), with 0 indicating that the program ran without error.
In the music thread, Petrushka wrote:
Petrushka’s point applies just as well to logical computer programs, and to our notion of error.
Judgements and beliefs are formed about objects (entities) in the world. In a correspondence theory of truth the truth of a judgment is seen in its relation to its object in the world, where the failure of correspondence indicates the judgement’s falsity (error) – at least that is my reading.
A few months back, KF admitted that he hasn’t read Royce, and that he was getting this argument from a Quaker theologian named Elton Trueblood. I haven’t read Trueblood, so I’m in no position to say anything about Trueblood’s interpretation of Royce
Royce does think that our recognition of error is immune to skeptical doubt. Differently put, he argues that any attempt to eliminate the conceptual possibility of error — “what would it be like if errors were logically impossible?” — results in a picture of our cognitive life in which we would be restricted to what Santyana called “the solipsism of the present moment”.
This is because error, as the mis-match between judgment and object, requires the possibility of a bridgeable gap between judgment and object. If there is no gap, then objects collapse back into judgments and we have nothing but our own judgments to steer by. If there is a gap, but it is unbridgeable, then again we have nothing but our own judgments to steer by. (Apparently Royce opposes naturalism/materialism — what he calls ‘realism’ — because he doesn’t see how the gap between judgment and object is bridgeable, under naturalism. It’s an interesting objection, though easily answered.)
I see the point here, but I think it misinterprets what Royce is doing. It is true that a false judgment is corrected by another, better, truer judgment — but Royce is interested in what happens before the judgment is corrected, when we notice that it is mistaken and in need of correction. And I think that Royce’s point would be more clearly stated if it were presented in terms of cognitive access instead of knowledge. Thus put, the problem is that the logical possibility of the recognition of error requires a form of cognitive access to objects distinct from assertions or judgments about objects. And that demand is easily satisfied by a correct theory of perception.
SophistiCat,
Yes, that’s right — Royce is assuming a correspondence theory of truth. But I don’t think there’s any alternative.
Coherentism is fine as a theory of justification, but it only works as a theory of truth by conflating the distinction between truth and justification. The so-called “pragmatist theory of truth” turns out to be, on closer inspection, the pragmatist method of conceptual explication applied to the concept of truth. Deflationary approaches just say that truth isn’t the kind of concept we need a theory about.
So if we need a theory of truth at all (and I’m not sure we do), it will have to be a correspondence theory. The real difficulty for pragmatic naturalists lies in re-conceptualizing both sides of the correspondence relation so that they are free of Platonic and Cartesian assumptions.
I think correspondence is fine as a semantic account of truth: that is, what we mean when we say that a proposition is true is that we have a “world” in mind, and the proposition is true if it coheres with our vision of this world, i.e. it corresponds to some feature of this world, as we see it. I am likewise dubious that any other account of truth is needed.
Likewise with error. What I suggested above was a semantic account of error, on which the realization of error is a divergence, a lack of correspondence, if you will, between our present judgement and some other judgment (our own at some point in the past, or someone else’s). This says nothing at all about being in possession of truth, in some abstract metaphysical sense.
SophistiCat,
On that account, doesn’t correspondence just collapse into coherence?
Strip away the verbosity, and KF is making the following fallacious argument:
1. Let S be the set of all statements that I, kairosfocus, consider to be self-evident.
2. Some members of S cannot be denied, because doing so leads to logical absurdity.
3. Therefore, no members of S can be denied.
I find KF’s verbosity so difficult to unravel that I would not hazard to translate his argument into numbered propositions.
As I said on thread about “self-evident propositions,” I don’t have any objections to the very idea of self-evident propositions. Of course there are self-evident propositions — I just want to understand what they are and how we come to know them. It is not self-evident what self-evidence is.
I was sorry to hear this, KN.
Maybe a bit of Lev Nikolaievich Tolstoy will help (e.g. “The Death of Ivan Ilych”), though he was excommunicated from the Russian Orthodox Church and not allowed back, even recently when his family tried for this. (Those darn Eastern absolutists must be to blame, in holding their views of Truth, right KN?)
If I were you, I wouldn’t spend much time worrying about UD or KF, it won’t help your career any and there’s not much worthwhile going on in the IDM these days, other than continued propaganda and also more and more science and faith conversations, which you don’t seem to be very fond of. Studying some Eastern philosophers might help, hmmm, resurrect your career, since the global balance of power is shifting.
Wilfrid Sellars, C.I. Lewis, Josiah Royce…the so-called ‘myth of the given,’ not exactly fascinating on the world scale, unless one is stuck thinking in the ‘washington consensus’.
In a ‘western’ language perhaps closer than Tolstoy to your stomach:
“if we need a theory of truth at all (and I’m not sure we do)…” – KN (professional philosopher)
I doubt even the ‘washington consensus’ would support your view of that.
Gregory,
Thank you, Gregory. In fact things have turned around for me a bit since I wrote that — I have a contract for my book and I have three chapters written so far. So I really can’t complain, apart from the abysmal state of the job market in the States for professional philosophy.
You’re welcome KN, and I’m sorry, but a book on ‘the myth of the given’ won’t save USAmerican pragmatism or the professional philosophy job market “in the States” where wisdom is currently at a discount. Too many people in the USA take it as a ‘given’ that the USA is a civilisation par excellence in late-modern global history that cannot possibly fall. The ‘washington consensus’ (even in philosophy, analytic, pragmatic, human cough) is a dying dinosaur on the world stage.
But then again, I might be (probabilities and statistics?) in error. There either is or is not some truth to what I’ve said above, and even more, in what Tolstoy said, which is apparently outside of your USAmerican analytic cum ‘continental’ philosophical interest to explore.
Nevertheless, on a subjective, relative scale, I’m glad to hear things have turned around for you and that you are writing, researching, creating, designing, dreaming…
Gregory,
Thank you, Gregory; that’s very kind of you to say.
There’s a highly oblique and deeply idiosyncratic line of thought that connects my professional activity in philosophy of mind/history of pragmatism — tracing out the vicissitudes of ‘the myth of the given’ — and my passion for social justice, environmental activism, and anti-capitalist movements (not that I’ve been doing much activism lately).
If I had to put into words what connects those themes, I would say it is a fully generalized anti-authoritarianism — I want to think about all norms or rules of ethical, epistemic, semantic, logical and political activity as ‘horizontal’ rather than as ‘vertical’. My ethico-political ideal is one in which self-organizing communities prevent themselves from becoming hierarchically ordered. (I suppose that, at the end of the day, I am an anarchist — though entirely unread in anarchist literature.)
I haven’t read much Tolstoy, but based on what my friends have told me, I suspect his Christian anarchism would appeal to me — at any rate, as much as it can appeal to a nice Jewish boy. But I have had a soft spot in my heart for Martin Buber ever since I read I and Thou during my late adolescence. I am a feminist and a socialist because I am a Kantian, but I am a Kantian because I am a Buberian.
The anatomy of error is a interesting thing since in origin subjects whole chunks of people on one side or both must be in error. Despite believing they are, and they are, quite sharp about thinking things through.
The whole point of science is to bring error possibility to its lowest possibility.
So is evolution a scientific conclusion or not or is its critics failing to be scientific or not./
Insofar as the claim is that normative facts — about what one ought to do or believe — are irreducible to natural facts — about what is or is not the case — I agree. If naturalism cannot accommodate the irreducibility of the normative to the natural, so much the worse for naturalism.
But I don’t think the prospect for naturalism are so dire as that. Just because the concepts of success or failure don’t apply to particle-exchanges or chemical reactions, doesn’t entail that they don’t apply to sentient animals living out very complex patterns of activities in their respective environments. As Allan Miller points out, a bird that flies into a window — or, better put, a bird that miscalculates the distance and angle of an approaching predator, and is eaten thereby — has made a mistake of some sort.
Now, it would be a philosophical mistake, I think, to say that the bird had made an error in Royce’s sense, because it made no judgment to begin with. There is a difference between being stupid (failing to adequately cope with objects in one’s environment) and being wrong (failing to adhere to social norms). (Dennett makes much of this distinction in his criticisms of Brandom.)
If the naturalist is not obligated to explain failures of proper cognitive functioning as shaped by past natural selection in non-sapient (non-rational) animals in terms of chemical reactions or particle collisions, then she isn’t obligated to explain failures of attributing and undertaking socially instituted commitments and entitlements to assertions in those terms, either.
However — and this is the precisely the key point — she might be obligated (on pains of abandoning naturalism) to explaining the game of giving and asking for reasons (attributing and undertaking socially instituted commitments and entitlements to assertions) in terms of proper cognitive functioning as shaped by past natural selection. I go back and forth on whether (1) that explanatory project is one that naturalism must shoulder, on pains of ceasing to be naturalism and (2) how feasible it really is.
I suspect that the concept of “error” will ultimately come down to matters of the synchronization of a nervous system with input from the senses.
Many people have experienced the disorienting effects of pain, high fever, sensory deprivation, gas poisoning, oxygen deprivation, nitrogen narcosis, narcotics, and anesthesia. Even near-dream states have disorienting features in which one is not able to distinguish reality. There is a point close to such disorientation where stability can be achieved by placing one’s ear next to the regular ticking of a clock.
All physical systems that are suspended in a state of uncertainty can go out of synchronization, but they can be brought back into a coherent state by some regular input. The more complex the system, and the closer to the edge of chaos it exists, the more it requires some sort of stabilizing input to keep it in a coherent state.
How that coherent state is manifested as a set of thoughts within living systems with complex nervous systems depends a great deal on the physical structure of its nervous system and its interconnectedness with the entire body of the organism as well as its sensory inputs from the surrounding environment. In this sense, synchronization with external input and cues constitutes reality for that organism.
In very simple systems – such as a bacterium following a chemical gradient or a heliotrope following the Sun – there is very likely no awareness of the difference between the internal state of the nervous system and the sensory input. Such systems survive natural selection by being “wired properly.”
But with sufficient hierarchical memory and memories of memories, such differences are perceived as “error;” and that “consciousness” of error can be further processed against other memories in order to determine what such an “error signal” means and how it should be used to change the internal state of the entire system. And, in such a complex system, there will be memories of the results of those decisions that will feed into the processing of future “error signals.”
The main upshot of such a “cold” analysis of a complex signal processing system is that such a system, left in isolation from its surroundings and left to interact only with other systems that are also isolated from surroundings, form a “super system” that can become desynchronized from external input. Such isolation provides no data for memory that would constitute the recognition of error.
So “error” in this sense constitutes – within a sufficiently complex, hierarchical memory system – the difference between what the system would be when synchronized by external input and its free-running, chaotic state with no input. Whether or not such a system recognizes such “error” depends on the hierarchical nature of its memory and the complexity of its structure.
Perhaps.
My point, in case it got obscured by side discussion, is that there is no problem here to be addressed, pace Royce (as I understand him, by way of your clear exposition in the OP). Let’s recall the premise:
Everyone agrees (it is self-evident) that people can err.
Royce then argues that by accepting this premise, we incur certain metaphysical commitments. I disagree. At most, accepting the premise obliges us to give a minimal semantic account of error: if we say that errors happen, what do we mean by that? And I think I gave such an account: at a minimum, what we mean is that our present judgement disagrees with some other judgement (such as our own judgement at some earlier time). This contains no metaphysical commitments to speak of.
Of course, people can mean more than that. Royce clearly has a certain metaphysical system in mind, which he links with the concept of error. He is welcome to it, but it is not strictly necessary to account for the premise of his argument.
Tolstoy can be rather grating when he gets into one of his philosophizing and moralizing moods, which he does often. A modern scientific wisdom, sometimes cited by teachers, is that every important idea or fact needs to be repeated several times (five times, to be exact) for it to sink in. Tolstoy was an intuitive teacher: he repeats his didactic points forcefully, time and again. He even does this with his characterizations: he picks some notable or idiosyncratic trait in appearance or behavior, and mentions every time a character makes an appearance. But what a powerful artist! My personal opinion is that his philosophy is… optional, but his art is for the ages. I would recommend Hadji-Murat – his last work of fiction; it is short, free of didactic commentary, and a perfect work of art, while containing astute observations, and strangely relevant in today’s world. It is also probably a less familiar Tolstoy, if you only read Anna Karenina.
SophistiCat,
I think the deepest part of your point, and one that deserves the most emphasis, is that one cannot get to metaphysics from semantics. (Alas, the huge cottage industry of “analytic metaphysics” consists of just that — trying to get to metaphysics from semantics or logic. It’s an intrinsically doomed enterprise, and I suspect it’s also one of the reasons why scientists take a rather dim view of metaphysics.)
So: if we interpret Royce as making a semantic point, then his metaphysics can’t get off the ground. (This is different from my criticism of Royce, which is that his argument for absolute idealism only appears to work because he hasn’t thought clearly enough about how perceiving is both similar to and different from judging.)
Now, to work:
On this proposal, which I understand to be a l analysis or explication of the concept of “error” — of the ordinary-language, “what do people mean by ‘error’?” variety — someone who says, “I thought that a snake, but it’s only a stick” is just contrasting two of her judgments — the old judgment, “that’s a snake” and the new judgment, “that’s a stick” — noticing that the judgments are incompatible, endorsing the new judgment and disavowing the old (but previously endorsed) judgment.
But, I think, Royce would still want to press the point — what accounts for the “that” in the two judgments? How do we know that the two judgments are about, or refer to, the same thing? Is the identity of reference between the two judgments a result of how they are combined in the mind of the person whose assertions they are — or does the person combine the assertions as she does because she recognizes that it is one and the same object that both judgments refer to?
Yes, that sounds right.
Well, Royce obviously doesn’t limit himself to making a semantic point, but his premise – the supposedly self-evident statement that he uses to get a foot in the door – necessitates only a semantic account, if even that*.
I don’t have the answer to this question. She may or may not have an answer. The question is: do we have to have an answer, in order to meet Royce’s challenge? I say, no. Whatever one happens to believe about reference, identity, realism/idealism, etc. – it is decoupled from the admitted fact that we make errors [of judgement].
* Arguably, not all concepts that are bandied around have a clear and universally shared meaning, when you dig below the surface. How about ‘free will’, or ‘identity’? One way to answer a challenge of the type that Royce poses would be to bite the bullet and admit that there is no agreed-upon account of the concept in question, and that people may mean different things when they use it in communication, or they may simply be confused about what it means.
SophistiCat,
OK, but Royce’s question isn’t “do we make errors?” but “how are errors possible?” — where “possible” already has, for him, the highly Kantian flavor of, “what the necessary conditions for the possibility of error?” — and, as I understand Royce, the question is better put as “what are the necessary conditions for the possibility of recognizing error as error?”
Royce does think that our recognition of errors survives hyperbolic doubt, and is self-evident in that sense, but he does so by way of a transcendental argument that a being that could not recognize errors as errors would have a cognitive life that consisted entirely of the solipsism of the present moment — in other words, nothing at all like the kind of being that we manifestly are.
(Since he does give a transcendental argument for the necessity of the possibility of error, he is not making a table-pounding “it’s self-evident, damnit!” gesture. A transcendental argument discloses the self-evident as self-evident.)
Also, notice that Royce’s interest in the necessary conditions for the possibility of recognizing errors as errors is different from the engineering question of how the basic parameters of the cognitive system thus specified are causally realized or implemented. (Consider that remark as a response to Mike above.)
I’ll be the first to admit that Royce’s starting-point for philosophical reflection is itself contentious — not everyone has the taste for transcendental reflection that I do! — and for me personally, the really interesting question is how to disentangle the starting-point of transcendental reflection from the idealistic metaphysics to which it is usually put — not just by Royce but in the whole tradition from Kant through Fichte and Hegel down to the “absolute idealists” of the 19th century.
The two — the philosophical question, and the “engineering” question — are deeply entwined.
Yes! But exactly how entwined they are is the very thing I’m trying to be clear and precise about!
KN,
It’s hard to imagine Dennett agreeing that being in error is the same as “failing to adhere to social norms”.
Dennett (arguing with and against Brandom) puts the distinction (somewhat jokingly, I think) as the difference between “being stupid and being naughty”.
The argument between them is complicated, but I think it turns on Dennett’s complaint that if we don’t start off with what is required for a simple cognitive system to successfully navigating its environment and then build up to the more complicated discursive practices that characterize human life, we run the risk of turning our account of the latter into a “skyhook” of givenness, and that’s incompatible with naturalism.
Brandom, by contrast, thinks that we can (indeed, must) start off with talking about the complicated discursive practices that characterize the life of normal mature human beings as rational beings, and only then develop more rudimentary versions of the same concepts in order to theorize about non-rational animals.
Aristotle says in his Metaphysics, when talking about what is “first,” that “what is first is said in two ways, what is first in relation to us and what is first in itself.” What is first in relation to us is what is most immediately at hand to us in our descriptions of our experience (the terminus a quo of philosophy or the ratio cognoscendi); what is first in itself is the complete account of the basic structure of reality (the terminus ad quem of philosophy or the ratio essendi). The progression from the former to the latter is the order of understanding — we begin with what is most evident to us and discover progressively more complex principles and truths — and the progression from the latter to the former is the order of being — we begin with the nature of reality and understand ourselves as part of it.
In those terms — and I know this is the right way to frame the debate between Brandom and Dennett because both of them are working off of Sellars, and this relation between the order of understanding and the order of being is central to his remarkable re-working of Hegel, who in turn got it from Aristotle — Brandom is defending the priority of the order of understanding over the order of being, and Dennett is defending the priority of the order of being over the order of understanding.
And of course that’s a misguided debate, because the order of understanding — which I understand in terms of transcendental philosophy — and the order of being — which I understand in terms of natural science, esp. cognitive science and evolutionary theory (the domains of natural science most directly relevant to philosophy) — have equal priority in contributing to an adequate understanding of the place of human beings in the natural world.
KN:
I’m not so sure. My example was related to the need of a flying animal to correctly and rapidly sense obstacles in ‘the real world’ and avoid them. I don’t think, by failing to distinguish glass from air, it was being stupid; it did make a judgement, albeit a very rapid, subconscious one, as we do every time we move or drive. And its judgement – that that part of space was OK to fly through – was in error.
The social norms thing, I don’t get at all. Perhaps it’s more of a scientific than a philosophical standpoint, but I’d see error as the distance of a measurement, perception, opinion, etc, from an actual state-of-affairs in the ‘real world’, regardless of how convention views things. It can be miles off – glass is not navigable – or a tiny bit.
You and Mike, I think, take error too narrowly. We recognize errors of abstract reasoning, of moral discernment, of judgement of character, even love (“I now realize that I never did really love her”). I think recognition of error is any disavowal of a belief, one’s own or someone else’s.
Also, just to get this out of the way, recognition of error does not imply an objectively better alignment of one’s judgement with respect to the world. No doubt, all of us have had an occasion where a “correction” gets us further off track. And not all of our judgments are closely linked to the world, anyway (such as above-mentioned abstract reasoning).
Kantian Naturalist,
I understand you saying that Royce’s account of error is building on some preexisting philosophical structure. However, I took the argument in your OP as a general argument that does not need any non-self-evident prerequisites (and surely it would be silly to deny that we make errors, in the usual sense). My point is that as an argument against someone not already sharing his metaphysical beliefs it doesn’t do anything: his question “How is error possible?” is easily answered, if we take the common-sense account of error as belief revision.
keiths:
KN:
Dennett is rejecting the idea that error reduces to violations of social norms. Here’s the full quote:
SophistiCat,
Even internal states and abstractions form part of the ‘real world’ which we detect. If we make a mathematical or rational proof, but get a step wrong, it is still ‘out there’ as much as ‘in here’. You cannot kick it, but you can externalise it, if only to the extent of simply chewing it over in the forefront of your thoughts (yes, I know, internal externalisation, incoherent, right?). If we think someone good but they are really bad, or revise our opinion on other previous feelings, we feel we made an error in the approach of prior opinion to some kind of ‘reality’: the accuracy of our perception of these less concrete sensations. Our sensations exist to us, even if we are the only ones who can detect them.
Not keen on that word ‘belief’. When one makes a measurement, one does not necessarily believe it to be accurate. One is aware of the limitations of the equipment, and the ability of the measured system to be informative to that scale. One can make repeat measurements, and converge on the ‘right’ answer, accepting that error frequently has a range, and is not dichotomous.
But that, as I say, is maybe heavily influenced by a scientific and computational background. As far as the fundamental capacity is concerned, for us animals particularly, error and the capacity to detect it is IMO intimately connected with learning. We need to recognise errors, as a fundamental survival mechanism. For next time, if there is a next time. It’s not learning from repeat error that is, in KN (via Dennett)’s terms ‘stupid’.
To externalize abstract reasoning would take some work. You can’t just use the reductionist’s stand-by of thoughts being “out-there” brain processes. Reductionism or not, brain processes are not the objects of abstract reasoning (when we reason about numbers, we don’t reason about brain processes that allow us to reason about numbers).
Anyway, my theme in this thread is that one should avoid unnecessary commitments. As a naturalist and (tentative) reductionist, I find your ideas about the “engineering” side of things plausible; I just don’t think that we need to haul this baggage in order to address the key question, i.e. how is error (logically) possible. The question doesn’t obviously demand a reductionist analysis.
But our thoughts are not limited to making measurements in the most scientifically rigorous and dispassionate manner.
“our thoughts are not limited to making measurements in the most scientifically rigorous and dispassionate manner.” – SophistiCat
Amen and behold possible elevation from your (tentative) reductionism.
Likewise, ‘naturalism’ is an unnecessarily narrow view (though R. Bhaskar does ‘naturalism’ credit) of existence that is easily cured outside of Anglo-American philosophies.
You are welcome, I’ll return to this someday soon…
Gregory, your holding my posts in moderation in the ‘Darwin’s Mistakes’ thread so no one can read my replies to you is one of the most cowardly shit-headed things I’ve ever seen.
This will get moved to guano but I want to make sure you and everyone else see it first.
Of course not, I was simply indicating a case where ‘belief’ feels too narrow. In all cases, for error to exist there needs to be something with which thought or action can be incongruent – a state-of-affairs of some kind which some perceptive mechanism can get closer to or further away from. Even with abstract reasoning, we assume a kind of ‘perfect’ rationale towards which we grope: the right answer’s in there somewhere. Or music: we have an idea what a ‘perfect’ rendition might feel like if we could do it. When we get closer than we were, sometimes things seem to click into place – the lightbulb moment.
One can have erroneous beliefs, but not all errors are a matter of belief – hence I prefer, more generally, some essence of incongruence with the ideal or the real.
Allan Miller,
I think you are making the same over-commitment as Royce does when answering a narrow question. No, we don’t have to assume anything about the correlation between our thoughts and states of affairs. A solipsist could still recognize an error (though, obviously, she would be limited to recognizing her own errors!) All that is logically necessary for a recognition of error to occur is that we recognize that our present judgement differs from some other judgment (e.g. our own past judgement), and that the difference is irreconcilable. Our present judgement is, by definition, our idea of what is true. If some other judgement is incompatible with our idea of what is true, then it has to be false: by our lights, the other judgement is in error. It’s as simple as that.
I’m more inclined to say that Allan is looking at it as a scientific question, rather than as a philosophical question.
keiths,
Thank you for providing the full quote. In retrospect it may have seemed that I was misrepresenting Dennett’s views — such was certainly not my intention!
My own pragmatism is closer to Brandom’s version — in which the account of error bottoms out in terms of the Community — than to Dennett’s version, in which the account of error bottoms out in terms of predictive utility. But a thorough-going pragmatism would integrate Brandom’s view with Dennett’s (and/or Millikan’s).
In particular, I am sensitive to Dennett’s worry that just appealing to the inferential norms of the Community risks turning the Community into a “sky-hook,” and I don’t want any more than Dennett does. And I find Brandom’s response to Dennett largely but not wholly convincing.
The most common mistake I see in “philosophizing” about stuff like this is that most of the people who do it have no deep understanding of science.
Science has grown out of a crucible of analysis of ideas, particularly about epistemology and ontology. This constitutes a long history that has culminated in activities that, to an outsider, appear to be simply mechanical and routine.
But the mere exercise of sitting down and writing a research proposal is an exercise that requires experience with epistemological and ontological issues. The vast majority of laypersons who have taken their required allocation of core courses in science have not received any exposure to any of this.
Regarding the concept of “error,” this concept has a clear, operational meaning in science and mathematics; and that meaning has emerged out of far deeper analysis than we are seeing in Royce and in the discussion taking place here.
Unless you really understand the notion of operationalizing a definition, such as the concept of error, then you have nothing to judge whether the concept you are dealing with has any meaning in any context. Such a word simply changes its meaning depending on context; and everybody ends up having to explain what he/she means in each context.
The definition of error as it is used in physics and mathematics is a far deeper and more complete concept than our local philosophers apparently understand here. I would recommend that, before this discussion drifts off into so many directions that nobody is able to keep track of where we are, our local philosophers study up on the definition of error in science, engineering, and mathematics. You may discover that this well-developed concept addresses more than you are aware.
SophistiCat,
I don’t find myself disagreeing strongly with that, so perhaps there is just something around the edges of our respective conceptions where we differ. Or perhaps not. I said this upthread:
Which seems similar, but for the interchange of judgement for perception. But perhaps the key distinction is my consideration of accuracy wrt some assumed external yardstick, vs your apparent argument that error lies in the difference between past and present judgements, independent of their correlation with any real or ideal referent? Surely, though, one can think oneself more right and yet be more wrong! The only way that can be so is if one has moved further away from something else, and so error must ultimately be ‘grounded’ in some way. Royce’s grounding is the ‘perfect Mind’, though there are obviously alternatives.