Realism and interfaces in philosophy of mind and metaphysics

In an earlier post, I discussed three theories of perception: naive realism, critical realism, and interfaces. To remind you of the terminology: naive realism is the stance that the world is exactly as we perceive it and critical realism is that perception resembles reality, but doesn’t capture all of it. Borrowing an image from Kevin Song: if naive realism is a perfect picture then critical realism is a blurry one. For a critical realist, our perception is — to move to another metaphor — a map of the territory that is reality; it distorts, omits details, adds some labels, and draws emphasis, but largely preserves the main structure. Interfaces, however, do not preserve structure. Borrowing now from Donald Hoffman: consider your computer desktop, what are the folders? They don’t reflect the complicated sequence of changes in magnetization in a thin film of ferromagnetic material inside a metal box called your hard-drive, not even at a coarse-grained level. Nor do they hint at the complicated information processing that changes those magnetic fields into the photons that leave your screen. But they do allow you to have a predictable and intelligible interaction with your computer, something that would be much more difficult with just a magnetized needle and a steady hand. The interface does not resemble reality, it just allows us to act. Although the comments section of the earlier post became rather philosophical, my original intention was to stay in the realm of the current scientific discourse on perception. The distinction between realism and interfaces, however, also has a rich philosophical history — not only in epistemology but also in metaphysics — that I want to highlight with a few examples in this post.

It is hard to find a strong philosophical proponent of naive realism, but I think Epicurus might be the strongest contender. Epicureanism focuses on sensation as the first criteria for truth — our sensations are not subjective and cannot deceive us about the external world that they attest to exist. Any falsehoods or deceptions enter into our experience only through the application of preconceived notions or categories to our sensation. These categories are fundamentally linked to language, and form the meaning underlying the word. It is tempting to equate this Epicurean empiricism with naive realism — in that the world is really like our sensation of it — but it is not clear to me that it would correspond to a theory of perception. Perception is not mere sensation, but also incorporates the Epicurean preconceived notions — once these are combined, we can systematically misrepresent the world, and thus have critical realism or even an interface theory, depending on the accuracy and source of our preconceptions. However, I think that the materialist nature of Epicurean thought has a lot of appeal to modern scientific readers, and so is a good historic source for naive and critical realism.

Surprisingly, you don’t need to be a materialist to embrace critical realism. In fact, you can build a whole non-materialist metaphysics around critical realism as Plato does vividly in the Allegory of the Cave. Our perceptions are like the shadows on the cave wall, created by — and in many aspects mimicking the properties of — the puppets that our captors set dancing to be illuminated by the flames behind us. Except for Plato, this metaphor is not meant to be limited to our perceptions — towards which I think his characters actually took a more naive realist view — but the whole material world, the objects of the world are just blurry photographs, imperfect maps, or shadows of the eternal world of Forms. These Forms give material entities their properties — like the Form of Beauty giving your significant other their beauty — and thus the material help us to remember the Forms — how studying your partner’s face helps you understand what it means to be beautiful — that your soul knew before the trauma of birth. An essential aspect of these forms — apart from their causal powers, which might seem particularly uncomfortable to the modern reader — is that many of the same properties can be applied to them as to the material objects that share in the Form. The sensibility of properties of perceptions in being applied to the perceived, is the hallmark of a realist theory.

Of course, critical realism was not confined to ancient Greece. It is alive and well today, and I want to touch an early twentieth century embrace — Russell’s (and to a much lesser extent, early Wittgenstein’s) logical atomism. Russell and the early analytic philosophers that followed him believed that language reflected the structure of reality. In this way, we could see them as picking up where we left off with Epicurus and being optimistic about the accuracy of our preconceptions. By studying language, we could arrive at the logical atoms that underlie meaning and through them understand reality. Of course, in the process we might learn that some of our sentences are shorthand for much longer logical statements, and that some sentences are even empty of meaning. If we follow the less optimistic Wittgenstein then we can even start to depart from critical realism, by seeing language as a veil that hides and distorts our underlying thought. In terms of Epicurus, by analyzing language we can come to discover in what way our preconceived notions fundamentally misrepresent reality, and thus arrive at an ineffable mystical understanding of the world. In my books, Russell’s optimist would clearly place him as a critical realist, but I can make less sense of Wittgenstein. For me, Wittgenstein’s direct mystical understanding of reality would qualify as a naive realism in its transcending of the Epicurean preconceived notions, but I don’t know how to classify his discussion of language (is it an interface to reality?) or if classifying it is even useful.

For full-fledged interface theories, although we can also find roots in ancient Greece — see for instance Parmenides‘s distinction between the way of truth and the way of opinion — I think the first sophisticated versions of this view originated in the late ancient and medieval times. The questions here were primarily concerned not with the material world but with the nature of God, and the main sources of evidence were not sense-data or experiments but syllogisms and scripture. However, similar philosophical tensions arose. For many philosophers, God was immaterial and had an existence that was distinct from that of the conditional existence of worldly things. Yet, infallible scripture spoke of God’s physical properties like hand, back, or throne or intellectual properties like power, mercy, wrath, and love. For interface theorists, these statements were meant as allegories, metaphors, and helpful presentations of God, but in no way reflective of God’s true nature. Rationalist thinkers like Averroes saw these scriptural interfaces as sent to be used by those who were not fortunate enough to dedicate themselves to philosophy. Sufi mystics like Ibn Arabi or Rumi extended these interfaces to a full metaphysics by suggesting that the whole of the universe was a veil disguising the one-ness of God. Like Wittgenstein much later, they would suggest an unveiling through transcending language.

Finally — and before you decide that I’ve abandoning all notions relevant to modern science — let me briefly mention Kant and quantum mechanics. To me, Kant is the most important interface theorist, for him the organization of our sense-organs create the phenomenal world as an interface with the thing-in-itself that we can have no direct experience of — kind of how I can’t really have a (meaningful) direct experience with the bits of magnetization in my hard-drive (or on the hard-drives of WordPress data centers) that ‘are’ this post. One of the most important structures of the phenomenal world created by our senses in this way is space and time. The conclusion that Kant arrives at — recast in Hoffman’s metaphor of a computer desktop — is that just because I can say that on my desktop the file for this post is to the left of the file for my most recent paper, I can’t say that the bits of magnetization that make up this post are to the left of the bits of magnetization that make up my paper (this can be further complicated by the fact that this post is actually on WordPress servers and my recent paper is on WriteLatex servers). Just because the interface has some structure, does not mean that this structure has to be a reflection of reality.

Kant’s viewpoint is important not only for philosophers, but appears in the interpretations of quantum mechanics. If you’ve ever heard of quantum mechanics, you have probably heard some form of Heisenberg’s uncertainty principle like “you can’t know both where you are and where you’re going”, or more formally: the position and momentum operators do not commute. If you investigate this facet of quantum mechanics as deeply as John Stewart Bell then you will see that any interpretation that is mathematically consistent with quantum mechanics will need to abandon realism or locality. Bohr’s interpretation — although made before Bell’s theorem was discovered — went the route of following Kant and abandoning realism. For Bohr, our experimental practices need to be grounded in pre-scientific concepts like space and time — in particular, the concepts of position and momentum are the formal specifications of these constraints of our phenomenal experience. Reality is not constrained in the same way as our perception, and so when we try to speak of nature in terms of our phenomenal categories — as we must in order to be able to communicate our experiences — we create the properties necessary for our interface. However, these properties — position and momentum, specifically — only exist in our interface with reality and not in reality itself. Thus, Heisenberg’s uncertainty principle is an expression of the tension between our perceptual interface and reality. Here perception is meant much more widely than human senses like eyesight and touch, and extends to all of our possible technological senses or experimental techniques: all of science serves as our interface.

Of course, this is not the only interpretation of quantum mechanics. We could also take the other horn, and give up locality as Bohm does, in order to preserve realism. We could also simply play with the meaning of ‘real’ and create three separate realities: manifest, scientific, and thing-in-itself; with the first giving us naive realism, the second — critical, and the third we can only interface with through the prior two. At that point it becomes tempting to declare “what is really real” between those three options to be a matter of taste.


About Artem Kaznatcheev
From the Department of Computer Science at Oxford University and Department of Translational Hematology & Oncology Research at Cleveland Clinic, I marvel at the world through algorithmic lenses. My mind is drawn to evolutionary dynamics, theoretical computer science, mathematical oncology, computational learning theory, and philosophy of science. Previously I was at the Department of Integrated Mathematical Oncology at Moffitt Cancer Center, and the School of Computer Science and Department of Psychology at McGill University. In a past life, I worried about quantum queries at the Institute for Quantum Computing and Department of Combinatorics & Optimization at University of Waterloo and as a visitor to the Centre for Quantum Technologies at National University of Singapore. Meander with me on Google+ and Twitter.

7 Responses to Realism and interfaces in philosophy of mind and metaphysics

  1. whitefrozen says:

    Im not sure if Wittgenstein was really about transcending language, early or late, so muchbas transcending philosophy. His early work is definitely concerned with a one to one language (representationalism is common to early analytic philosophy) but he recognized that lots of things didnt fit into his scheme (ethics, aesthetics, etc), hence his famous statement on silence. His later, meaning-is-use work was a repudidation (sp?) of his earlier work, and he realized that language fundamentally works without being one to one.

    Kant, of course, has the unfortunate quality of basing his philosophy on physics, which had the consequence of his philosophy lasting as long as the physics he based it on – which isnt to say he didnt have sound insights (but just as many not so sound insights).

  2. Hello Artem, here is my very drafty critic. If the discussion grows, I can add more with more links. Here is it:
    This is an interesting post by , there are many things to write about and the discussion could go on for hours, but I’ll try to be short and shoot the gist of my critic.

    I start by saying that I like interfaces more than naive or critical realism. But this does not mean that I believe interfaces are qualitatively better than other forms of realism.

    I’ll get to it after I describe a view from the plane of Artem post.

    Here is it:
    – … naive realism, critical realism, interfaces
    – epicureanism
    – Plato as non-materialist (realism and materialism are not the same thing)
    – the Allegory of the Cave ( examine the Greek versions of these two key paragraphs
    to discover that “reality” in the 515c and “real” in 516a are artifacts of the biased translator, when truly in the Greek version is used “unconcealed”. Perhaps this comment is not even worth to pursue, because too subtle, for example the fact that there is an IF at the beginning of the Allegory, and there is nowhere discussed anything else than what are the relative views in and outside the cave, therefore nowhere “reality” as understood by empiricist appear there)
    – use of the analogies naive realism=photograph, critical realism=blurred photograph
    these itch me badly because when it comes to perception these sorts of analogies lead straight to the homunculus fallacy. Marr comes to mind , reminding about the crevice between cognitive science and CS (the first has the right negative stance that any photo analogy is just crap, but does not have yet the right tools to get enough quantitative results, the second can get quantitative results at will but has an untenable philosophical position which has plenty of manifestations in the limitations of the solutions they propose for the problem of biological vision, see , a classic respected by both sides, about that )
    – … and we get to Russell, Wittgenstein, the usual analytic philosophy. Don’t get me wrong, analytic philosophy is great, but the usual confusion I see almost everywhere in CS people with a philosophical bent, is to think that empiricism is a kind of correct form of platonism, which is of course obviously bogus. Nevertheless Artem announces here the color: the a priori is empiricism. Which makes me cringe: if we start from there then there is no problem concerning reality, end of story.
    – Parmenides, great, but what about Timaios, would not that be a better source?
    – … and the usual Western path of interpretations of ancient Greeks via the Arabs, which is after all the source of the misunderstanding and empiricism=platonism without waving hands dubious point of view (another long discussion).

    Now, if anybody survived the plane view, let’s now go to second thoughts. Dennett argument called the Cartesian Theater springs into mind as more relevant than Plato’s cave for this discussion.
    Where do naive realism, critical realism and interfaces stand in Dennett theater? Is clear: the spectator who believes that the characters on stage are “real” is naive, the one who is in a state of suspended disbelief during the show is a critical realist and the professional critic sees the show as the interface between the intentions of the director and the public.
    Of course that, among these three ways of experiencing the show, the professional critic one is the most elaborate, but fact is that, excepting the rare naive realist, no one believes that is, or needs to suppose any (iso)morphism between yet another greater stage of realism and the show. (Here is the homunculus fallacy again, you cannot escape it, darn, there must be because of the empiricist stance.)
    Going back to Artem article, there is also Kant and quantum mechanics. I pass.
    … and I go back to an interesting point which has not been stressed before in this critical post. Is the analogy between the interface point of view on perception and the folder system on a computer. It is a nice one, I reproduce the paragraph: “[the folders] don’t reflect the complicated sequence of changes in magnetization in a thin film of ferromagnetic material inside a metal box called your hard-drive, not even at a coarse-grained level. Nor do they hint at the complicated information processing that changes those magnetic fields into the photons that leave your screen. But they do allow you to have a predictable and intelligible interaction with your computer, something that would be much more difficult with just a magnetized needle and a steady hand.”
    So, I understand from this analogy that:
    – the silicon and whatnot computer is like reality
    – the computer user is you (oups, the homunculus again, darn)
    – and the folder system is the interface
    Is a great analogy, even putting aside the homunculus stuff. But I can’t forget that the folder system is the distant but faithful descendant of the Porphyrian tree, which is a tool invented by Aristotle to put some order in the human knowledge (read: hash table).
    Therefore let me put together what I could get from Artem article, but mind that perhaps this is not exactly what Artem wrote or mean:
    – 1. take an empiricist stance
    – 2. and assume that perception is like an interface from an indisputable reality and one’s brain
    – 3. and the interface is a semantic classifier

    May be good for a computer system.

  3. Hi! I finally had time to read the post, so here is what I thought.

    Let’s look at the notion of “reality” is an explanation vehicle. What does the it explain? If the interface is between postulated reality and our mind and it doesn’t preserve structure, what does it preserve of reality? A reality we know nothing about besides its mere existence is not going to help us explain anything. I believe we should avoid bringing up notions of which the only sure thing is that they exist. It is not coincidental that God is brought up in this post in the history of interface theories section. The interface metaphor framed as an interface between reality and our mind suffers from the same problem as religion does. Although religious thought went behind this actually, in the “Mystical Theology” of Pseudo-Dionysius or, say, Nagarjunas thought the notion of reality itself is seen as interface, allowing Areopagitus to say “don’t believe those who say god exists”, the implication being the critique of our human notion of existence. God to him is not existing, but not not existing.

    The notion of interface is only then useful, when we can know something that lies on both sides of it. If the only thing we ever get is the interface – interface is our reality and all the information we get, all the notions we invent do not transcend it. They are only effective as categorizations and classifications of the processes in the interface.

    We don’t know everything about reality, but does knowing more make what we already know more real? I don’t think so. My understanding is that in the process of acquiring knowledge we are discovering new information about reality, and, arguably, creating some reality.

    In short — I can think about my senses as an interface between my mind and my environment, but not as an interface between my mind and reality, because reality is more then just environment.

    • There’s no way to edit the comment, so sorry for a couple of misprints! (is=as in the first sentence, there should be no the in the second, then=than in the last sentence…)

    • I like the perspective you offer here on reality, and the important distinction you make between reality and environment. I am also partial to the Epicurian perspective of ‘reality is what you sense, and you cannot be wrong about that’; this is how the manifest and scientific images get formed. I think that at this point it becomes important to become explicit with respect to what modalities, we are discussing ‘reality’. It can be “real to me”, “real to society”, “real to particle accelerators”, “real to reason”, … I also really appreciate the historical context, and the discussion of existence as a category itself (thus susceptible to all the critiques we are offering against other categories).

      The only line I want to disagree with is:

      The notion of interface is only then useful, when we can know something that lies on both sides of it.

      I am not sure if we need to know what is on both sides. We can learn certain features of interfaces simply by studying our minds, while having a non-interface theory would force us to always study the environment. Of course, this is complicated greatly by evolutionary accounts of the mind, since parts of our ‘interface’ through evolution can absorb certain structures of the environment (or our fitness function, more specifically).

      However, there are definitely certain ‘regularities’ that arise from how our interfaces or theories are structured, and not from the environment. Abel discussed one example, recently, but my favorite would be Noether’s theorem. We spent a long time being amazed at energy conservation and ‘testing’ it experimentally. We even called it a law, until we figured out that it was just a mathematical artifact of us assuming that physics is symmetric with respect to time. In this case, it might have seemed like energy conservation was a property of our particular environment (that could have, for instance, been false in another similar environment), but then it turned out to be a mathematical consequence of how we defined our concepts of energy and measurement and the implicit assumptions we made along the way. We could discover this fact (i.e. Noether theorem) without knowing “what is on the other side”, by simply studying our interface (the theory, categories and terms we defined). Of course, when we are working outside of the context of formal theories, this becomes much more difficult and hand-wavy, but that is just a typical day for philosophy.

  4. Pingback: Cataloging a year of blogging: the philosophical turn | Theory, Evolution, and Games Group

  5. Pingback: A year in books | Theory, Evolution, and Games Group

Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s