Hesperus is Bosphorus

A group blog by philosophers in and from Turkey

Epistemic Realism and the goal of Epistemology

with 21 comments

I’m teaching a class this Semester on ‘The British Realist Tradition from Reid to Williamson’, and I tell my students that I’m an ‘Epistemic Realist’ and this post is an attempt to work out what I mean by this. Anyway here’s a first, inadequate, stab at explaining what I mean by epistemic realism: “There is such a thing as knowing, and one central goal of epistemology is to understand more clearly what sort of thing it is”.  I think that knowledge is something like a mental natural kind (or perhaps a set of distinct natural kinds) and the task of epistemology is not primarily to get a better understanding of our concept of “knowledge”, but to discover truths about knowledge and to provide a better conceptualization of this aspect of the mental. A central question for an epistemic realist has to do with the relationship between our epistemic language and epistemic facts – and on this I’m sympathetic to Thomas Reid.

Following Reid: (1) I’m a believer in the defeasible authority of common sense. (2) I think that it is not immediately clear what belongs to common sense and what does not. And (3) I take the fact that a certain distinction is found in all natural languages to be a defeasible indication that the distinction is part of common sense. Thus I think that if a distinction is to be found in all languages, this is a good indication that it reflects a real distinction in the world. (I discussed this briefly in a previous post here)

Reid argues that every language contains a word like ‘know’ and that, “[T]he verb know, denotes an operation of mind. From the general structure of language, this verb requires a person; I know, you know, or he knows. But it requires no less a noun in the accusative case, denoting the thing known; for he that knows, must know something; and to know, without having any object of knowledge, is an absurdity too gross to admit of reasoning.” (EIP p.44)

Now, as far as I am aware, all natural languages contain a word that can be translated by ‘know’ and in all the languages I know this word is a transitive verb that requires both an active subject and an object. [One question I have is whether in all languages one can say “we know” – does anyone know of a language where one needs a different word for this?]. I take this to be defeasible evidence that knowing is what Reid calls a [kind of] mental operation that requires both an agent and an object.

The problem with appeals to the agreement of natural languages on this point is that different languages carve up mental space slightly differently. So, for example, in Turkish the word for to know (bilmek) is related to the word for information (bilgi) and, perhaps as a result of this, in Turkish it is more common than in English to talk of ‘false knowledge’ or ‘knowing wrongly’. (For a discussion of this see a recent paper by my colleague Murat Baç and Nurbay Irmak here). In German there is a distinction between erkennen, kennen and wissen, which can all be translated as ‘to know’. Kennen is normally used to express knowledge of an object, whereas wissen is used to express something closer to knowledge of a fact. So, for example, one would translate both (1) “Kennst du den Autor des Buches?” and (2) “Weisst du den Autor des Buches?” into English as “Do you know the author of the book?” The first German sentence is asking whether you are acquainted with the author of the book, whereas the second one is asking whether you know who the author is.

What are we to make of such linguistic difference? One common thought is that the words “know”, “bilmek” and “wissen” refer to slightly different concepts. If we think that the task of philosophy is conceptual analysis, then epistemology in English and epistemology in Turkish will have slightly different subject matters. As an epistemic realist, I reject such relativism. In using the word ‘know’ we are attempting to refer to a mental natural kind, or set of natural kinds, and I think some languages do a better job than others. So, for example, I suspect that the German distinction between kennen and wissen probably captures a real distinction between kinds of mental states or attitudes, and that it would be useful to adopt such a terminological distinction into English. Of course English speaking philosophers have made similar distinctions, but we haven’t got simple terms to capture this distinction, so my take here is that the English verb ‘know’ is probably ambiguous in a way in which German is not. For example, as we all know, Russell famously argues for a distinction between knowledge by acquaintance and knowledge by description. And a distinction like Russell’s is to be found in many earlier British philosophers. So, for example, Hobbes distinguishes between a certain type of knowledge (and error) that is possible without language (which he calls “knowledge of fact” and includes the knowledge acquired by sense and memory; this knowledge is “the knowledge required in a witness”), and a distinct type of knowledge, which he calls science, which he thinks is made possible only by the introduction of names.

My thought here is that although common sense, expressed in our languages, doesn’t carve up nature at the joints, it gets pretty close – and at least as a starting point it’s the best guide we’ve got. And some languages are better butchers than others. I think that English speaking philosophers in the 18th and 19th centuries had a much richer active vocabulary for talking about the mental than we do today, and that for various reasons the analytic tradition led to an impoverishment of our vocabulary for talking of the mental. I suspect that many of our works for talking about the mental are ambiguous, not picking out natural kinds, and one of the aims of epistemology is to enrichen our vocabulary – and one way of doing this may be to look at how other languages, and philosophers in the past, did this. So, for example, one distinction that was ubiquitous in 18th and 19th century British philosophy was the distinction between belief and opinion (and between believing and opining). I think this is a useful distinction that captures a real distinction but which has been largely lost to contemporary philosopher. And I think it might be useful to reintroduce it into our philosophical lexicon. Hopefully I’ll post on this some time soon.

I’m worried that this post sounds very naive and unsophisticated. But appeals to common sense often do. One thing worth pointing out is that for Reid there is no real hard and fast distinction between a priori philosophy and empirical research.

There was recently a discussion on NewAPPS on the language requirement for philosophy phds (here). I guess that the sort of points I’m trying to make in this and my previous posts may suggest one motivation for requiring a language requirement for philosophers of language and epistemologists.

UPDATE: Here’s a link to Murat and Nurbay’s paper that is not behind a paywall: “Knowing Wrongly: An Obvious Oxymoron, or a Threat for the Alleged Universality of Epistemological Analyses?

Written by Lucas Thorpe

March 11, 2012 at 8:51 pm

21 Responses

Subscribe to comments with RSS.

  1. Re: distinction between believing and opining. Are you familiar with Dennett’s discussion of this in his “How to Change Your Mind?” (reprinted in Brainstorms, 1978)? He sees opinions as dependent on language and close to what most philosophers have meant by “belief”, while he argues belief is whatever rationalizes an agent’s total pattern of behavior. This enables him to explain akrasia as acting on one’s beliefs though against one’s opinions. Not sure if that’s close to the distinction you have in mind, but it’s at least one example of a contemporary philosopher trying to revive it. Nice post. Love Reid, even though I think common sense is woeful at representing the natural facts about the human mind!

    Tad Zawidzki

    March 11, 2012 at 9:26 pm

    • Hi Tad,
      Thanks for the response and kind words.
      I haven’t read Dennett for a long time – so wasn’t familiar with his discussion of the distinction between believing and opining. Thank’s for the pointer – I’ll have to take a look. When I have a bit of time I will post something on the 19th century distinction between belief and opinion. I guess looking at the options that were available then might create space for a slightly richer language in discussing the mind.

      There are a couple of things I like about Reid. I think he was perhaps the most attentive and observationally acute of the modern British philosophers. he is, for example, probably the early modern philosopher who has the most interesting things to say about developmental psychology – his observations about the importance of imitation in development, for example, are extremely perceptive and relevant today.
      I also think that many philosophers today think that common sense philosophy is opposed to a more scientific attitude towards the mind – but I think that this is a mistake. Reid distinguishes between two major branches of philosophy – the philosophy of the material world, which he calls natural philosophy and the philosophy of the “intellectual world” which he calls “Pneumatology” – which I guess would include, amongst other things, epistemology, philosophy of mind, ethics, developmental psychology and (today) psychology. the big hero of natural philosophy is, of course, Newton. And his understanding of pneumatology does not give priority to armchair speculation. Of course, he is not aware of the breakthroughs in the methodology of modern experimental psychology – but I’m sure he would have embraced them. I think that a lot of contemporary philosophy of mind and epistemology is extremely unrealistic when it comes to the taxonomy of mental states and attitudes. And I think that a more realistic epistemology should appeal both to common sense and empirical research. And I don’t think that there is any real tension between common sense and empirical research. Indeed I think that the starting point of any empirical research has to be common sense (although, obviously, this is not necessarily the endpoint of such research). And a richer common sense taxonomy of mental states might be useful for empirical research.

      Lucas Thorpe

      March 11, 2012 at 11:23 pm

      • I love Reid on testimony, and his appeal to natural, developmental facts. And I love his direct realism. His refutation of Hume’s argument for representationalism anticipates ecological psychologists like Gibson, and notions like the direct perception of affordances, I think. However, I must disagree with you on the continuity between common sense and the sciences of the mind. Perhaps you’re right that we must start with common sense. (Where else?) But I think common sense, especially as it’s encoded in public language, has a very different goal than science. It’s aim is ideological – encoding regulative ideals that make possible group life. It is not scientific – describing human nature as it is. I think Hume’s right, for example, on the notion of a persistent self – it’s a myth useful for assigning blame, rather than a referring term. And I would extend that sort of view to most folk psychology. But these are just my idiosyncratic peeves.

        With regard to cultural variation, though, what do you make of language that seem incapable of even attributing propositional attitudes (e.g., Piraha)? If common sense is a good guide to human psychology, then what to make of the common sense of the PIraha? On my view, linguistically-encoded common sense is useful not for epistemic but for socio-regulative reasons. So cultures that have no normative need to encode commitments to complex propositions needn’t represent propositional attitudes.

        Tad Zawidzki

        March 12, 2012 at 3:59 pm

        • Hi Tad,
          Thanks for the response. Do you know of a good reference (a starting point) for the Piraha language and propositional attitudes? As you may have noticed from previous posts Ilhan and I are very interested in propositional attitudes.
          I think that the relationship between common sense and science is complicated. I do think that the science of psychology is in its early days and we are only starting to scratch the surface of the mind. And so I suspect that working our more clearly the common-sense ontology of the mental can play a useful role in the development of experimental psychology. So I take common sense and science to be on a continuum rather than opposed. I also think that in philosophy the ultimate methodology involves inference to the best explanation – and so I think that scientific results do and should play an important role in philosophical arguments.
          In terms of ordinary language, cultural differences and experimental psychology I think there is some interesting work being done. My friend, Emre Ozgen, for example, in the psychology department at Bilkent is doing some interesting work on basic colour terms and differences between different linguistic groups. (Turkish seems to have one more basic colour term than most other languages – I wonder if my Turkish friends know what this is?)

          Lucas Thorpe

          March 12, 2012 at 4:13 pm

        • Hi Lucas. Actually, there’s a piece in Biology and Philosophy (2007) by John Bolender (who I think teaches somewhere in Turkey) about the philosophical implications of what is known about the Piraha. Probably some good references in there. I’m also very interested in linguistic differences and their effects on cognition. There was recent work done (by I forget who) on an extra word for a shade of blue in Greek, and the consequences of this for color discrimination. Gumperz, Levinson, John Lucy, Nisbett, and many others are all doing very interesting research into cultural and linguistic relativity. Also, Marc Slors, in the Netherlands, is engaged in research regarding cultural variation in folk psychology specifically. I recently reviewed a grant application he made to the Dutch government to research this. That grant was denied, but he’s found other sources of funding apparently. All of this is Googlable. I don’t have time right now to track down the precise references (sorry)!

          Best, TZ.

          Tad Zawidzki

          March 12, 2012 at 4:20 pm

        • Hi Tad,

          John Bolender taught at the Middle East Technical University for many years. He is now in Canada – and just joined this blog as a contributor. I’ll have to ask him to send a copy of his paper.

          You mention “There was recent work done (by I forget who) on an extra word for a shade of blue in Greek, and the consequences of this for color discrimination. ” If you switched Turkish for Greek here this would describe some of Emre’s work. e.g.

          Ozgen, E; Davies, IRL. 1998. Turkish color terms: tests of Berlin and Kay’s theory of color universals and linguistic relativity. Linguistics 36 (5): 919-956.

          Ozgen, E. 2004. Language, learning, and color perception. Current Directions in Psychological Science 13 (3): 95-98

          Some links can be found here:

          Lucas Thorpe

          March 12, 2012 at 4:27 pm

        • Hey Lucas. Thanks for those. Here’s the paper I had in mind:


          Tad Zawidzki

          March 12, 2012 at 4:32 pm

        • Thanks for that. 🙂

          Lucas Thorpe

          March 12, 2012 at 4:35 pm

        • Here’s a link to the paper by John Bolender (mentioned by Tad) on “Prehistoric cognition by description: a Russellian approach to the upper paleolithic” in case anyone was interested.

          Lucas Thorpe

          March 12, 2012 at 5:04 pm

  2. Lucas what was the distinction between *believing* and *opining* in the old days? Why do you think the English language lost the verb *to opine*? I am also thinking of the difficulty in translating the old Greek term “doxa” into English. Is that at all relevant here?

    ilhan inan

    March 11, 2012 at 9:29 pm

  3. Re Ilhan:

    Hi Ilhan,

    I don’t know if this will interests you but I think there is quite a well defined modal distinction between to opine (meinen) and believe (glauben) in Kant –I am sure Lucas will have more to say about this.
    Briefly, to opine, to believe and to know (wissen) are three stages of holding something (i.e., a proposition) to be true (Fuhrwahrhalten). Accordingly, to opine is to entertain a proposition as its mere POSSIBILITY without making a commitment to its truth. This kind of judging a proposition, Kant also calls, problematic judging. Kant thinks that we opine when there is neither objectively, nor subjectively sufficient evidence for holding a proposition’s true.
    To believe is stronger attitude of holding a proposition ACTUALLY true or judging it assertorically. Yet, believing is short of knowing in that it only has subjectively sufficient grounds but lacks objective grounds. That is to say, we are subjectively committed to the truth of the proposition, yet we are also aware that there is not enough ground for the truth of the proposition in objective reality. (Belief in God is a perfect example.)
    To know, is of course, not only subjectively but also objectively grounded commitment to the truth of a proposition. The interesting (and hard to digest) thing is that Kant adds that “knowing” comes with a consciousness of the necessity of the truth of the propositions. Hence, to know is to judge apodictically.

    There are tones of complications of this modal account (not many Kant scholars are willing to accept this connection between modalities of judging and epistemic attitudes), but I think this is what Kant does.


    March 12, 2012 at 9:38 am

    • Hi Uygar,
      Thanks for the comment. The person here who knows most about this is actually Berna Kilinc – she’s done quite a bit of work on this distinction in Kant. The distinction between believing and opining in the British tradition was a different from this. One difference, I think, is that belief seems to be something that comes in degrees, whereas opinion does not.

      Lucas Thorpe

      March 12, 2012 at 11:43 am

  4. Excuse my typos. This was the first time I posted a comment 🙂


    March 12, 2012 at 9:40 am

  5. Hi Lucas,
    I am a bit late joining the debate. One question I have pertains to the notion of natural kinds operating in your epistemic realism. Are you committed to natural kinds essentialism, or would a definition of natural kinds as homeostatic property clusters be sufficient for you? As I see it, at its best, the literature on natural kinds is messy and the concept of natural kinds does not individuate any feature of the natural world without qualifying what one takes natural kinds to be.

    Serife Tekin

    March 14, 2012 at 5:56 pm

    • HI Serife,

      I’m not really sure what I think here. I guess that I think that there are mental operations (or acts or states – I’m not committed here) and that an operation of believing is a different kind of operation than an operation of knowing. So I guess I’m committed to something like a belief in natural (mental) kinds. But I’m not really sure whether or not I’m committed to (mental) natural kind essentialism. I haven’t looked at the natural kind literature for a while. But from what I remember it is kind of messy. I guess I might be happy with a fairly messy account of natural kinds as clusters (although I’m not really sure how to characterise such clusters). I guess I might be committed to some basic similarity relations. And perhaps kinds can be loosely individuated in terms of such similarity relations rather than by appeal to some essence.

      But – I’m a believer in the division of intellectual labour- so perhaps I can just appeal to natural kinds, assuming that the notion of mental natural kinds makes sense, and leave others to provide a reasonable theory of natural kinds. But I’d certainly like to have a more satisfactory grasp of what natural kinds are. Perhaps I don’t need to appeal to natural kinds; perhaps there is another notion that would serve my purposes better. But for the time being this is the best (although vague and unsatisfactory) way I’ve found to think about what I mean by being an epistemic realist.

      I guess we’ll talk a bit about this when you’re here in the summer. Do you have any suggestions about the best recent literature on this? Do you know any good recent survey article, for example?

      Lucas Thorpe

      March 14, 2012 at 6:18 pm

      • Hi Lucas,
        Division of intellectual labor is certainly helpful; I might have a recommendation: One way to get around the difficulties associated with the messiness of the literature is to use paradigm examples of what you take natural kinds to be. If what you are seeking is merely similarity relations that hold together otherwise logically unrelated and non-accidental individuals (which I think would be a non-risky move) you might want to give a paradigm example, like something from the animal species, e.g. cats, and work with that. People like Ian Hacking, Rachel Cooper, Muhammad Ali Khalidi have chosen this strategy –though with different motivations– and I think it is helpful.
        Richard Boyd’s account of natural kinds as homeostatic property clusters (HPC) might also be helpful. It feels like HPC might capture what you have in mind.
        Though not recent, in a 1991 volume of Philosophical Studies Hacking and Boyd has a very good exchange, which offers a good summary of the main challenges that face natural kind essentialism. Here is the link, let me know if you can’t download them. I can send them to you:
        In the more recent literature, Muhammad Ali Khalidi has good papers on the topic, which are accessible through his academia.edu page:
        Rachel Cooper has a couple of chapters in her 2007 book “Psychiatry and Philosophy of Science” but they are more specific to natural kinds debate on mental disorders. They may not be as useful.

        Serife Tekin

        March 14, 2012 at 6:43 pm

        • Hi Serife,

          Thanks for the links – one problem with this blog is that I now have a growing number of paper is my ‘to read’ pile. But I’ll definitely have a look at this literature – and I guess we can talk about it when you come. I guess I need to work out what I think here more carefully.

          I guess we’ll try and have a weekly reading group over the summer – so perhaps we could look at some of this stuff. (or some other literature).

          Lucas Thorpe

          March 14, 2012 at 6:54 pm

        • Having lots of things to read is a good problem to have. Reading these together would be great. I am finishing a paper on NK now, which hopefully will settle some of my anxieties around it. I might have more to say afterwards.

          Serife Tekin

          March 14, 2012 at 6:59 pm

    • Wearing my Kantian hat for a moment: I think that at least one kind of mental operation might need to be understood in essentialist terms – the bringing stuff (for want of a better word) under the transcendental unity of apperception. but this would be compatible with regarding acts of knowing, believing etc as belonging to kinds in some looser sense.
      And I’m not sure if I really want to wear my Kantian hat – perhaps this depends on how unified I feel when I wake up in the morning

      Lucas Thorpe

      March 14, 2012 at 6:31 pm

      • Ha ha! Kantian hats are useful for mind jogging. The helpful feature of the HPC is its amenability to be used by natural kinds essentialists as well.

        Serife Tekin

        March 14, 2012 at 6:45 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: