The "internal/external" metaphor in the philosophy of mind

Helge Malmgren

PhD, MD, Associate Professor
Department of Philosophy, Göteborg University, Sweden

Poster presented at the conference Toward a Science of Consciousness ("Tucson 2000"), Tucson, Arizona, April 10-15, 2000


1. The Cartesian intuition in epistemology
2. "In me" and "in my mind"
3. Abstract and concrete senses of content terms
4. The visuo-spatial metaphor
5. Internal sense and reflective introspection

1. The Cartesian intuition in epistemology

Even though Cartesianism in epistemology has been challenged by many, the thesis that the subject’s own consciousness and its contents are known to the subject in a special, intimate and secure way remains a living option in philosophy. However, if one searches for rational arguments for this thesis, there seems to be very few.

One possible such argument is that reconstruction of Descartes’ thought in which it is claimed that since "I exist" follows logically from "I doubt that I exist", it is certain that I exist. But few contemporary philosophers think that this deduction can carry any substantial epistemological burden. Of course the problem now turns out to be, "How do I know that I doubt?" Neither are Descartes’ arguments from the existence of a benevolent God any longer believed by anybody to be valid.

Another class of arguments (the immanence arguments, for short) for Cartesianism in epistemology start with the observation that in cases of knowledge about one’s own mind and its contents, the objects of knowledge are in the subject and/or in her consciousness and/or in her knowledge-act, while this is not so in cases of knowledge about the so-called "external world". It is then usually taken as obvious and in no need of further arguments that "inside-ness" entails a special epistemic status. Here is an illustrative quotation from that very influential 20th century Cartesian, Edmund Husserl:

The being of the cogitatio, or to be more exact, of the knowledge-phenomenon itself, cannot be questioned, and it is free from the riddle of transcendence. /.../ It is also obvious, that the cogitationes represent a sphere of absolute immanent givens, in which sense we also interpret immanence. In the seeing [Schauen] of the pure phenomenon, the object is not outside the knowledge, outside consciousness, and at the same time it is given in the sense of the absolute self-givenness of a purely seen object."

E. Husserl, Die Idée der Phänomenologie (Nijhoff 1958, original lectures 1907), p. 43. My translation and underlining.

My main question is now, "Why does the fact — if it is a fact — that our mental contents are in us, or in our mind, or even in the knowledge-act, make our knowledge of such contents certain?". To answer this question, I first have to dig rather deeply into another one, namely, "In what sense, if any, are my mental contents in me, or in my consciousness"? My answers to these two questions will be:

(1) None of the senses in which my mental contents are "in" me or "in" my consciousness makes it at all probable that my knowledge of these contents is more certain than my knowledge of the so-called "external" world;

(2) The idea that mental contents are known in a special way just because they are "inside" mainly derives from the use of "in" as a visuo-spatial metaphor.

(3) The use of this metaphor is partly due to a confusion between reflective introspection (the observation of one’s own mental states) and internal sense (which provides knowledge about one’s internal bodily states).

2. "In me" and "in my mind"

In what sense, if any, is my mind "in me", and in what sense, if any, are my mental contents in my mind?

To start with the latter question, let us first observe the valuable distinction drawn by Husserl between intentional and real contents of consciousness. In the weak, intentional sense whatever I happen to think of (for example Pegasus, or the car which I see) is in my mind. Nobody believes that such "intentional immance" entails any prileveged epistemic status, so this sense of "in" can in principle be left out of the following discussion. But there is also a stronger sense in which my present feelings are in my mind, but Pegasus and the car are not.

Philosophers who defend Husserl’s distinction usually hold that one’s conscious acts are also in one’s mind in the latter, strong sense — they are real contents — even if they have intentional contents which are not. However, it may also be the case that an act has another real content of consciousness as its intentional content. This, says Husserl, is importantly true in immanent perception, which according to him is the main source of our privileged knowledge about ourselves.

So, the first kind of content relation which is essential to the Cartesian position in epistemology is that of the mind’s real contents to the mind itself. Another one is in play when a philosopher says that the mind is in or within the person, or even inside the body. In a third, derivative sense the mind’s real contents are sometimes said to be within the person. In the following quotation from Locke, all these three relevant uses of the content terms are illustrated.

Secondly, the other fountain, from which experience furnisheth the understanding with ideas, is the perception of the operations of our own minds within us, as it is employed about the ideas it has got; which operations, when the soul comes to reflect on and consider, do furnish the understanding with another set of ideas which could not be had from things without: and such are perception, thinking, doubting, believing, reasoning, knowing, willing, and all the different actings of our own minds; which we being conscious of, and observing in ourselves, do from these receive into our understanding as distinct ideas, as we do from bodies affecting our senses. This source of ideas every man has wholly in himself: and though it not be sense, as having nothing to do with external objects, yet it is very like it, and might properly enough be called internal sense. But as I call the other Sensation, so I call this REFLECTION, the ideas it affords being such only as the mind gets by reflecting on its own operations within itself.

J. Locke, An Essay Concerning Human Understanding (Oxford 1924; the first edition was published in 1690), pp. 43-4. My underlinings.

It is important to note that Locke uses ”internal sense” for what I prefer to designate as ”reflective introspection”, and not for what I (with most contemporary authors) would call ”internal sense” (see Section 5 below). Also note that the inside phrases — "our own minds within us", "in ourselves", "into our understanding", "in himself", "internal sense" and "within itself" are here contrasted with the outside phrases "things without" and "external objects". I will in the following refer to both kinds of terms as content terms.

3. Abstract and concrete senses of content terms

Let us first note that content terms are used in a large variety of situations, which have only certain formal properties in common. An egg can lie in a basket, but the tones are also in the melody, and there can be an error in a mathematical argument (while the mathematicians’s sexual preferences are outside the relevant mathematical considerations). More generally, the part-whole relation in the formal, mereological sense holds between any entity and a complex whole of which it is a part. Hence it is a very abstract creature and cannot possibly, as such, be of any great use for a Cartesian epistemologist. With this I mean to say that if she wants to argue from the fact that a sensation is in her, or in her mind, or in her act of immanent perception, to the conclusion that she knows some things for certain about this sensation, she must fill the word "in" with more sense than the formal mereological one. Which specific sense, then, can carry the required burden? Perhaps the ordinary, spatial senses of the "inside" and "outside" terms could be rallied to do some job in epistemology?

A. But remember that G.E. Moore, when attempting to prove that the external world exists, did not hesitate to use his hand as a paradigm example of an external object. Hence what is called "the external world" in epistemology is not spatially external to us, since it includes our own bodies and their parts. Likewise, introspection is not inspection of what is spatially internal. If it was, we should be said to introspect when we looked into the interior of our stomach with a gastroscope. Our mind and our mental contents are not spatially interior to us or to our bodies. Neither are our mind’s contents spatial contents of our mind. (This, of course, also follows from Descartes’ thesis that the soul is not an extended substance.)

Hence even though spatial containment often entails an epistemological advantage to the spatial contents in questions (cf below), this fact cannot be of any value for the Cartesian epistemologist. My argument on this point is strengthened by the following consideration.

B. We often think of and imagine our minds as being spatially interior to our bodies, and mental contents as being spatial contents of the mind. One could argue that this fact shows us a phenomenological but still quasi-spatial sense in which our minds are in us, and in which the mind’s contents are in the mind; one could further argue that it somehow gives a rationale for Cartesianism in epistemology. But it does not:

B(I). A spatial representation need not represent anything spatial. We often represent numbers as being spatially laid out on a line, but we would not dream of drawing the conclusion that certain numbers really are to the left of others. The only thing we can say with some confidence is that a representation in terms of spatial contents most probably represents some relation with the same formal structure as the spatial content relation. But then we are back in the abstract mereological sense.

B(II). If our spatial representation of the mind really represents a belief that the mind is a spatial entity, this arguably expresses a grand illusion. Since the mind is not spatially in the body (cf A ), any apprehension of it as being so is essentially false. How, then, could the existence of such an apprehension be useful as a premiss in epistemology? (I do not say that it cannot. I just ask how.)

It seems that other senses of the content terms than the ordinary, spatial ones must be involved in any rational reconstruction of Cartesian epistemology from "inside-outside" premisses. But which could these senses possibly be? Here are two suggestions:

C. The mind and its contents are constituents of the complex union of body and mind, or of mental and physical events, which is called "The Person". Hence they are interior to us (as persons) in a concrete sense, and this sense confers an epistemological privilege to them.

Regrettably, exactly the same argument can be carried out with the body and its parts in the central role. Hence the essential epistemological asymmetry (for the Cartesian) between mind and body cannot be defended by way of this line of reasoning.

D. Being a real constituent of a consciousness is a relation sui generis which has no counterpart in the physical world. It is this very relation which is the basis of our privileged access to our own mental states.

My only objection to this suggestion is that although it may well be valid, it would be utterly dogmatic to believe in it without further arguments. Some such arguments can be found in the writings of contemporary Cartesians such as Husserl. However, I cannot discuss most of Husserl’s detailed claims here (as you may guess, they do not convince me at all). Let me just add that it is not at all self-evident that all parts of all knowledge-acts are known by the knower. Hence Husserl’s thesis that the object of knowledge is sometimes not only (really) in consciousness, but also in the knowledge-act itself, does not relieve him from explaining why this immanence should entail an epistemic privilege.

4. The visuo-spatial metaphor

There are certainly several other possible senses in which it could be rationally argued that the immanentness (being-inside) of a mental content makes it epistemologically privileged. However, because of space limitations I will instead — in this and next section — go into what I think are the main irrational sources of Cartesianism in epistemology.

If you are indoors, looking around yourself, your field of view is totally or almost totally limited by a set of irregular surfaces at varying distances from you. The possible exceptions are some holes and/or transparent surfaces through which you can look at Infinity in the form of the clear sky; let us disregard these exceptions for the moment. This visually connected (but physically usually disjoint) set of limiting surfaces I here call the visual room. Let us call an object an interior visual object if one of its surfaces belongs to the visual room. Now the following propositions are true about your visual room and the interior visual objects:

1. The interior surfaces of the visual room are the only surfaces which you can see directly;

2. You see the interior visual objects more directly than you see any other objects;

3. You are yourself an interior visual object.

The reader is urged to contemplate the structural similarity between propositions 1-3 and the basic tenets of Cartesian epistemology. This similarity is my first argument for the thesis that Cartesianism is often nourished by a visuo-spatial analogy.

The second main kind of evidence which I would like to bring forward is the abundant use of visual terms among certain Cartesian philosophers, such as Husserl. Anybody who has read Husserl extantly knows that his use of "seeing" (Schauen) in the quotation above is not an isolated occurrence!

5. Internal sense and reflective introspection

Of course, I cannot even try to prove my main thesis here. But I want to add a few more considerations, blocking a certain kind of counterargument. Why, says my imagined opponent, should we at all be tempted to use a misleading spatial metaphor when thinking about our knowledge of the mental? Of course it is fact that we tend to use spatial thinking in a lot of contexts — but as you yourself said above, we are usually not mislead. Why is thinking about self-knowledge different? Here is my answer:

Each one of has relatively privileged access to the interior of his/her own body. We receive signals from receptors in our own joints, but not from those of our fellow-people; likewise we can feel that we are hungry or satiated in a way in which we cannot feel that our fellow-people are. Although much of the knowledge which derives from the "internal senses" does so in an unconscious and implicit way, many of the signals from these senses do become conscious. That is, they deliver conscious internal sensations, informing us about the status of our bodily parts and processes.

Now the epistemological privilege which goes with internal sense in this sense is a contingent and relative one. It is just a coincidence of nature that there are no nerves leading from your joints to my brain. Furthermore, an "outside" observer in the form of a surgeon with a gastroscope is sometimes much better suited to judge the status of your gastric surface than you are (without the gastroscope).

But yet the privilege is a privilege, since in many situations you are the only one who can tell anything at all about the interior state of you body. My simple suggestion is that the existence of this privilege may make you believe that the way you know your own mental states is essentially similar to internal sense. And if you believe that, you may also be mislead to think that your mind is a spatial entity with an interior, just as your body is. This is the confusion between internal sense and reflective introspection which I have already hinted at.

It is of course not difficult to see that the two are different. Taking our departure from Husserl’s version of reflective introspection, the transcendental reduction (bracketing the world), we can take the study of one’s perception of a tomato as our paradigm example of reflective introspection. Such a study is not at all inwards-directed in the way attending to one’s heartbeat is. But although the distinction is obvious when brought into the open, I am convinced that many philosophers suffer from the confusion on a subconscious level.


This work was supported by a grant from The Swedish Council for Research in the Humanities and Social Sciences (HSFR). For helpful comments, I am grateful to Anna-Sara Malmgren, Erik Olsson, Filip Radovic and Susanna Radovic.

Top of page
Back to Papers and Posters