One last (hopefully) brief post on Externalism. My complaints about some of the posts over at On Philosophy didn't take. Peter still wants to critique externalism in terms of causality. So let me raise a few of the issues of why various kinds of externalisms take the approach they do. Further let me outline why I don't think this is ultimately an issue that can be resolved by empirical means.
The first obvious example is contextualism. In semantic matters this is roughly the idea that the context of a sentence is implicitly tied to its context. The most famous example is Putnam's twin earth thought experiment. There we have two substances phenomenally identical but one is water whereas the other is something else. The question then becomes do people on each world mean the same thing when they say "water." Now in terms of their internal states they are identical. But because on one world what they refer to isn't water we have an intuitive sense that there is something different.
Note, however, what is going on. First off this isn't an empirical question about the brain. Rather it is a question of what we mean by meaning. That is one can ultimately see it not as an empirical question but a question of our language. You could make every empirical claim you wish, but the externalist will agree upon every claim about the brain. The debate therefore is whether meaning (or more accurately semantics) is all internal. As Putnam puts it, "meaning just ain't in the head." Note though that this doesn't involve causality per se. (Beyond the claim that the causes to a person's perception can be identical yet reflect radically different substances) Rather it is a claim about what our words refer to.
The second example is epistemological externalism. Now this comes in different forms. Alston and Plantinga offer variations on reliabilism for instance. Roughly though we can associate the various forms in the claim that an external observer can look at a person and say they know or don't know but that this judgement doesn't purely relate to what is present to the person. (That is what they are conscious or potentially conscious of) This is opposed typically to the traditional internalist form of epistemology where we have justifications that the person knows of that enable him to form a justified judgment putting him into a state of knowledge.
Now to be fair while philosophers often talk of epistemology it doesn't get discussed quite as much in philosophy of mind. What is considered center isn't knowledge but belief. Knowledge is simply a particular kind of belief that happens to be true. That is it is fundamentally seen as a subset of belief. (Typically by associating it with justifications) The problem is, of course, that "truth" can't easily be formed in a sensible way in terms of pure internalism unless one moves to a radical form of idealism. Thus correspondence. But note what this does. Knowledge, as true justified belief, excludes something from the interior - the truth condition.
It is not at all unusual to consider "knowing x" to be a mental state. But if it is a mental state and depends upon something external then internalism simply can't handle our language. But once again note what is at stake. This isn't an issue of causality nor empirical issues. Rather it is an issue of what we mean by knowledge and whether our common language takes knowledge as a mental state. Fundamentally it is an issue regarding the meaning of "mental" or "mental descriptions." What is to be allowed or disallowed.
For this reason externalists with regards to epistemology tend to provide arguments based upon fundamental intuitions regarding our use of words. (Timothy Williamson's recent Knowledge and its Limits being the best example - although I find many of the arguments in it problematic or unpersuasive)
Note that the issue isn't whether the externalist can martial sufficiently persuasive arguments. Rather the issue is about what is at stake. And that is, I feel, the meaning of knowledge, the meaning of meaning, and the meaning of mental. While I'd definitely never argue there aren't empirical implications I think that ultimately the issue is linguistical or taxonomical. That is what is to count as knowledge or meaning and what the limits of these terms are. (I'd note that the opening chapter in Williamson is an extended defense that knowledge is in fact a mental state.)
Now I'm not here trying to defend any particular kind of externalism. There are literally dozens of kinds of externalism. But what they all share is that the extension of some term can't be limited to physical states within the nervous system. Some might say they depend upon what is external to the mind, with no assertion regarding what the mind is. (I think Putnam's semantic externalism is an example of this) Others simply say the mind is more than the brain. But at a fundamental level it is a debate over the meaning of all these terms and what their limits are. To misunderstand this is, I feel, to misunderstand the debate.
I made some of the above points and related discussions in my post about intentionality and my post about externalism the core issue. One ought perhaps check out the Wiki on externalism which does a great job briefly outlining the main positions. The SEP on externalist justification is very good for the epistemological issue. I'd also check out their articles on externalism about mental content and their discussion of the brains in a vat thought experiment. At the end of their entry on intentionality is a section on externalism. Their entry on perception contents also has a nice section on externalism.
I should note that when the externalist and internalist debate do have empirical overtones such as in cognitive science, the issue is less over what the terms mean in common speech than in terms of what cognitive science should take them as. (i.e. should the terms be narrowed) At least that is how I read papers such as Chomsky's that argue semantic knowledge is a purely internal state. The issue is less what our common sense knowledge of what "Bill knows he is in water" means than what the cognitive scientist is interested in.
This to me though establishes the point I was making earlier. An externalist might very well completely agree that cognitive science should focus purely on internal computation but disagree that this is what our terms mean in philosophy. (Although some might argue that to understand cognitive functions one must fundamentally understand the relation of the person to their environment with those relations being types and not individual causes)
I'd note that Chomsky wisely deals with this by talking about E-languages and I-languages. The former are social and thus open to externalist analysis whereas the latter are narrow in content and more focused on what he is concerned with. I think thinkers like say Peirce who fundamentally see knowledge and inquiry as a social activity of a collection of people and not the individual would agree. Thus we have to distinguish this sense of the social from the individual. Once again though the externalist would have no problem with this. They'd simply critique the limits of the internalist project but wouldn't have any problem with the analysis itself. Just that something essential to our understanding of semantics, language, and so forth is excluded by the analysis.
Did you read this? I am also reminded of Husserl's Logical Investigations Volume 2 Part 1 (German Edition), investigation 1, where Husserl shows how you can construct the idea of objective truth and an objective meaning for words form an internalist position (he does this by distinguishing meaning from sense, the distinction I have called objective meaning and personal meaning). Have you read his work? I am curious then as to why you think that Husserl fails.
Peter we're leaving for the hospital so I just don't have time to say much and probably won't for a week. I have read Husserl (although its been a while). My objections end up part and parcel of the objections of most of the Heideggarians. I'll see if I can't address it some time in the future. But I find Husserl's notion of objectivity problematic.
Just to note, I'm not primarily here focusing on why one ought be an externalist. In the analytic tradition, as I think I said, the strongest arguments are by Burge and Putnam. But I don't think there is a knockdown argument either for or against externalism or internalism. At frankly a lot of the arguments both for and against are rather poor (IMO). I'd add that the related problem is the wide variety of positions within each side. For instance I put up an article on the sidebar I believe yesterday that focused on whether Physical externalism and Semantic externalism are compatible given that one relies on a reference theory of types and the other on a descriptive theory of types.
That's why I cautioned lumping all externalism together. While there are some similarities between all the kinds of movements the differences are extremely important. Sometimes even working out what kind of externalism a philosopher is logically committed to is difficult. (Consider for instance the debate over semantic and social externalism in Heidegger) Likewise some externalists, like say Davidson, are outspoken critics of some forms of externalism.
So the situation is much more complex than the simple internalist/externalist label suggests. Largely the debate is over the following:
1. what counts as mental properties or descriptions
2. where mental properties are instantiated
3. what individuates particular properties or descriptions
There are many possible ways to answer each of those. As I mentioned in my post, for instance, Williamson's epsitemological externalism is committed to knowledge and not just belief or belief + justification as counting as a mental property. So there's a lot of debate there. I take the Burge/Putnam debate as primarily focused on (2) although this in turn requires particular (and perhaps incompatible) answers to (3).
But I'm not saying internalists don't have reasonable positions. Just that it isn't quite clear from a linguistic point of view whether they are satisfactory to deal with the kinds of descriptions we give in our natural languages. That's why I find Chomsky's approach so welcome. He acknowledges what might be a fundamental equivocation in natural language and just separates the issues for empirical investigation.
Oh, one last bit I should add in. While the dominant view of Husserl is as an internalist clearly he demands context be part of what we typically term meaning. There is an ongoing debate about whether the interalist reading of Husserl (largely, I believe, influenced by Herbert Dreyfus' reading) is correct. I'll confess as to not being up on the details of this debate. But as I understand it the question is whether Dreyfus' reading of Husserl in a Fregean way. So there is a real debate about whether Husserl is an externalist, internalist or (as say Dan Zahavi argues) something else.
This "something else" that arises out of the notion of the noema is interesting and is something I ought investigate further but I've just never had the time. (I want to carefully go back through Derrida's critique of Husserl first and Davidson's thought on language and Frege -- I'd started that this spring but haven't had the time to continue it like I wish. So only a few posts made it out of the readings.) My sense is though that this non-Fregean reading of noema ends up just entailing a particular kind of externalism that is different from say the Putnam kind of semantic or physical externalism.
And of course those issues are also tied to whether one reads Husserl as an idealist or a realist.
The think that implicitly bothers me about any discussion of justified true belief is that many of the examples and counter-examples seem to treat a justified true belief as a vaguely justified belief, but not a true belief.
In a non-purely-analytical context I don't see how any belief can be considered to be *completely* true unless it is infalliable, with of course a large spectrum of beliefs that turn out to be largely true, or partially true.
So I would say that knowledge is knowledge to the degree that it actually is true, i.e. to the degree it corresponds to things as they are, as they were, or actually will be - things in this case being the domain of the real, the objective, and the existent, (things you can be wrong about) depending upon the precise metaphysics.
Now the whole basis of a realist conceptualism, like that of Ockham, is that there is an infinite continuum of conceptual flexibility, that thinking involves discovering and fixing on a definition or archetype of a particular concept in mind, and then truth of propositions was based on the correspondence between the conceptual structure and the actual external objects under consideration, and that this correspondence came in degrees, according to an objective similarity measure between concept in mind and object or objects in reality.
So this skips the whole problem of whether any particular concept is real, as such. Ockham would simply say that concepts are objective. Similarity, in the most general sense was the replacment for real, discrete universals - it encompassed all conceivable universals on a continuum derived from the real similarity between objects (you can see that for Ockham similarity was an extremely rich 'concept').
So on an Ockhamist account, the two world problem is trivially resolvable - he would say that there are no real formal distinctions, that a distinction in reason only, is not a real distinction at all. Everything about the difference between XYZ and H2O is captured by the real world similarity between the two, not the way we think about it, or which concepts we adopt to describe it, or any properties not directly present in XYZ or H2O itself.
Any distinction in the truth value of a particular proposition could either be a distinction of convention (e.g. a formal or conceptual distinction) or a natural distinction (according to the real (external) difference between objects), and in practice a grand conceptual hierarchy with natural signs typically on the bottom and synthetic concepts largely built out of natural signs on the top, but nonetheless carrying truth value, in terms of the degree the synthetic concept corresponded to the objective or external world.
Very neat stuff, if you ask me. Here is an interesting paper that I am sure is rather more terminologically accurate than my description:
Russell, Strawson, and William of Ockham
Sharon Kaye, Dalhousie University
http://www.bu.edu/wcp/Papers/Medi/MediKaye.htm
Heh, I actually went to Dalhousie University. (My Dad was a professor there and I later transferred to BYU)
I don't have time to carefully read your comments. Something strikes me as off. That is the issue of whether similarity implies no real difference.
I also don't think one can say a belief has to be infallible to be true. I can believe weakly that my wife is upstairs. It seems that if she is upstairs it is true regardless of whether my belief is infallible. But I'm probably missing the point you are making.
I'd add that Gettier examples make the whole approach of justified true belief as knowledge perhaps problematic. Of course some dispute this. I kind of like Williamson's approach (if not his arguments) for making knowledge a foundational mental state. But I think his approach is definitely controversial and its unfortunate more of his arguments weren't stronger.
My own view is that perhaps knowledge is one of those vague and equivocal terms we use in natural languages. I am quite sympathetic to Peirce's tendency to avoid debates about knowledge entirely and merely deal with beliefs that are fixed. (i.e. that we can't as a real practical matter doubt, regardless of whether they are true)
While I think investigations into evidence and justification are tremendously helpful, as too is the focus on correspondence, I think that philosophers tend to push these concepts further than is perhaps appropriate.
Getting back to correspondence, so long as we don't take correspondence the way it is typically taken as foundational in analytic philosophy, I don't have a problem. As I've often mentioned I rather like the way Peirce talks about it as what the ideal community of inquirers would believe if they inquired sufficiently far. I think that takes care of equivocation too (although not all agree) since one need only say that the community has discovered all the possible choices even if it is ultimately undecidable. This is a kind of conceptualism, although certainly not the traditional one (and I think not really compatible with Ockham although clearly Peirce was influenced by Ockham)
The issue then becomes what is the relationship between a concept in this sense and concepts in the sense that I have them. After all the difference between my concept of me typing on the computer right now and a parallel version of me doing the same in the Matrix seems conceptually different even if I can't tell them apart. I think Peirce can capture this but I'm not sure Ockham can.
Clark, I believe I addressed some of those issues. A particularly critical one is Ockham's concept of similarity by degrees, and not just linear degrees either, but similarity as a real relation that encompassed the full real (external) semantics of the similarity and difference between two objects.
In short, similarity is not an either/or question to Ockham, but encompasses everything that can be said about an object. His idea of natural signs or concepts did not require a complete familiarity with members of a species or genus, nor the idea that species and genera were everlasting ideals, just a sufficient natural acquaintance with objects of the appropriate kind as to naturally distinguish the concept in ones mind from other concepts.
[Note: That could be a slight exagerration - I would have to research whether Ockham considered identicality to be equivalent to maximal similarity. However that is a question of nomenclature, that is easily resolvable either way.]
A material point is that concepts as a whole are potentially fuzzy, i.e. there could be a broad spectrum of horses that are more or less horse-like, where the constraints on what is meant mentally by the word horse could either be natural (based on familiarity with actual horses) or conventional (based on restricting what is considered to be a horse to a formal list of qualifiers).
The point is is that he believed the external world generally speaking had no 'idea' of what a horse was. There was no eternal ideal of a horse, nor was there an essence of horse-ness that inhered in all actual horses. The only real 'thing' that horses had in common was similarity, conceived as a rich, real two place relation between any two objects that contained all the information about the similarities and differences between the two, and that this relation existed in the outside world, completely independent of all minds. ('Real' for Ockham means existent in the outside world, roughly speaking, 'Objective' refers to concepts that refer, by degrees to existent or possible realities).
But back to your main issue, Ockham's concept of similarity includes *all* the properties of an object, including location. He holds to the principle of indiscernability of identicals, where properties are constrained to be real properties, distinctions of reason are not real to him.
Now similarility comes in many other forms and degrees, for example he described in terms of his ontology what it means to be members of the same species (Roughly that members of the same species share all the essential properties of a referent of a concept).
According to Marilyn Adams, Ockham held the following beliefs about similarity:
1. Similarity is a real relation that, given the existence of the relata, holds independently of any will- human or divine-and hence of any conventions
2. Similarity comes in degrees ranging from trivial to maximal and including generic and specific similarity.
3. What make it the case that certain particulars are co-generic or co-specific is that they are similar genus-wise or species-wise.
4. Whether certain particulars are co-generic or co-specific is neither arbitrary nor subjective.
The way Ockham arrives at (3) and (4) is his theory of natural signification.
5. Natural signs are concepts that signify by by virtue of real relations tht obtain or not independently of any will-human or divine.
Ockham says:
"...Similarity is called a real relation because of the fact that one white things, by the nature of the thing, is similar to another white things, and the intellect no more brings this about than it brings about that Socrates is white and Plato is white."
Now it is very important to recognize that none of this works properly with a strictly bivalent logic - too many Sorites paradoxes - to Ockham whether an object was a representative of a concept was a matter of degree. The critical difference between his theory and "Ostrich" nominalism, is his doctrine of natural signs.
The paper I linked to above addresses this pretty well - that similarity between objects in the outside world, even conventional similarity as manifest in history (e.g. the concept: "the King of France") is more than adequate to give such concepts meaning whether the referents now exist or not.
The critical thing is that meaningful concepts be objective, not subjective in terms of evaluation (but not necessarily construction). Anyone can devise a new synthetic concept, make it manifest in the real world by building examples of it, and now the concept is fully objective, not imaginary or subjective any more - i.e. the described similarities exist in the external (real) world, not just the world of the intellect.
I'll read that when I get back from the hospital. I'm leaving right now. My apologies for reading your post so superficially. I've been in a rush. It sounds very intriguing although I'm not at all convinced one should make "similarity" real. But it is very intriguing.
To add, I have learned from sad experience that the medievals, perhaps moreso than any other class of philosophers, require careful reading and thought before commenting on. Every time I've quickly whipped off a comment dealing with medieval philosophy I've regretted it. (You might remember the embarrassing site of me conflating the two Scotus in a discussion at T&S on the different senses of Being) They are all very subtle, especially Duns Scotus and Ockham.
Clark, Ockham says that similarity is not a *thing*, but that it is both mind-independent and in the outside world, and that makes it real by his definition. Concepts, on the other hand, he does not consider in the realm of the real, but rather in the realm of the objective, or at least potentially objective.
The most direct mapping between reality and the objective is the concept of a natural sign, which reflects an apprehension of the patterns, species, genera etc of the outside world, in the inside world in sort of a mental language.
One of the reasons why similarity is not a *thing* is that everything in some category is trivially similar with everything else, i.e. similarity is a real abstraction, in a manner similar to the abstractive reality of all true mathematical theorems.
Similarity doesn't have any material existence of course, it is just a relatively natural characteristic of things that causes them to be apprehended as natural species.
By the way, I hope it is apparent here that Ockham is using a rather different sense of the word natural here, then I am when I speak of natural law. Natural here simply means "existing in conformity with the observable world" (Wordnet), whether the species is of absolute, ordinate, human, or incidental origin.
I understand Ockham to avoid the term real species, because he is agnostic on whether there is such a thing, and convinced that most species are not real, but rather a conceptual artifact or reduction of certain types of similarity. So he says that yes there are natural species, but there is no law or essence out there strictly defining where one species starts and another ends.
I get the impression from the discussion in Adams that half of his problem with essences is the requisite discreteness for them to be things, not to mention the strange idea that all ideas have corresponding essences in the natural world, especially essences that subsisted in the instantiations of ideas.
Sorry, by "thing" I didn't mean to imply reification of similarity. (Bad for me since I make the same distinctions in Heidegger and Peircean discussions) More that it was mind-independent whereas I'm not convinced it is. I was actually musing on that a month or so ago and am still thinking through the issue.
Of course the notion of real iconic signs is the big problem with my musings on similarity. So it's definitely an area I'm thinking about. And, since to be frank, iconicity is for me probably the key issue in philosophy along with the issue of repetition of signs or token-type relations it is something I've been thinking about a long time and still haven't come to fixed conclusions. (Indeed those two issues probably determine the philosophers I've studied the most)
Clark,
I think it is much of Ockham's point that something (say a sign) being natural does not necessary imply that it is real. Ockham does not believe that universals are real, as a general rule, but he certainly believes that many concepts are natural.
The difference here is that natural signs or concepts (in mind) occur on an infinite continuum. Some of them might be more faithful than another, and indeed which concepts are more faithful can be a time varying question, due to the changes in the real world objects they are used to abstract.
No for various reasons, I am half inclined to believe in at least a small handful of real generals. However, I think that in the large, Ockham's picture is more accurate than any schema that has a plethora of real generals. I just cannot see any particular evidence for such a large number, except in the most abstract terms, like the theorems of mathematics.
Mark, you've convinced me to read up on Ockham. Knowing how difficult the original sources are (although I constantly refer to them), what in your experience is the best book on him. I've read things on him but nothing terribly sustained or comprehensive.
Clark,
Most of my detailed knowledge of what might be termed Ockham's analytical philosophy is from William Ockham, Volumes 1 and 2, by Marilyn McCord Adams, University of Notre Dame Press, 1987, 1402 pages total.
Adams does not treat Ockham's moral, ethical, or legal philosophy in any great detail in those volumes. My understanding of that is based largely on collected articles of a more recent date.
The two volumes are divided into five parts:
I. Ontology
Nine chapters on the problem of universals, names and concepts; universals, conventionalism, and similarity, quantity, relations, quality, and an evaluation of Ockham's ontological program.
II. Logic
Three chapters on the properties of terms, the logic of propositions, and arguments
III. Theory of Knowledge
Two chapters on conceptual empiricism and direct realism, and certainty and skepticism in Ockham's epistemology.
IV. Natural Philosophy
Six chapters on the metaphysical structure of composites, matter, quantity and individuation, intensification and reduction of forms, efficient causality, motion, and time.
V. Theology
Eleven chapters on divine simplicity, faith and reason, God as a knower, divine ideas, divine power and possibility, extent of divine knowledge, omniscience, human freedom, and future contingency, divine omnipotence, theologism, grace and merit, and predestination, reprobation and freedom, human and divine.
I have only read about one fourth of this material so far. One of the great strengths is the level of treatment of the competing programs of Scotus, Aquinas, and others, and Ockham's arguments against them. It is worth noting that Ockham considered Scotus "the Subtle Doctor, who excels others in subtlety of judgment", and gave his arguments the most detailed treatment.
I presume he deals with the engagement between Ockham and Scotus and the places where their presumptions are different? The Cambridge guide I have does a pretty good job with the famous cases, but when I read about a philosopher I often like to read "the other side" which some treatments neglect.
The answer is a definite yes - many of the chapters (especially the early chapters) start with an outline of the historically preceding positions that go on for several pages before dealing with Ockham's specific counter-arguments and his own positions.
[Marilyn Adams is a she, by the way]
I've closed comments in order to avoid spam since I don't check this older blog as much anymore.
and Theology
Number of unique visitors:
Blogged by Clark Goble