Michael Huemer, Skepticism and the Veil of Perception

Chapter VII:  “An Objection to Indirect Realism:
The Problem of Spatial Properties”

1.  An Argument against Sense-Data

    In this chapter, Mike Huemer offers the following argument against sense-data:

“1.  In perception, the things I am directly aware of (at least sometimes) have locations.

2.  Only physical things have locations.

3.  Therefore the things that I am directly aware of in perception (at least sometimes) are physical things.”  (149-50)

2.  Comments on this Argument

    The crucial premise in this argument is premise 2, and, even leaving aside the question of sense-data, it immediately gives rise to a number of questions, including:

1.  Do human minds have locations?  If they do, then, according to premise 2, human minds must be physical things - presumably brains.

2.  If human minds do not have locations, then what account is to be given of the fact that what happens in one particular brain is causally connected with one particular mind, and not with other minds?

3. Do mental states - such as sensory experiences, thoughts, and feelings - have locations?  If they do, then, according to premise 2, such mental states must be physical things - presumably certain physical states of relevant neuronal circuitry.

4.  If mental states do not have locations, then what account is to be given of the relationship between physical states of the neuronal circuitry in one particular brain and the mental states of one particular mind, and the absence of such a relation between physical states of that neuronal circuitry and the corresponding mental states of other minds?

    In addition, many religious people, and some present-day philosophers, believe that human minds are immaterial substances.  There seem to be good empirical objections to this view.  But even if it is not true that human minds are immaterial substances, it is surely true that humans could have had minds that were immaterial substances.   But if premise 2 is not only true, but necessarily so, then it would follow that such immaterial substances would not have had any spatial location.  But why would it be impossible for such immaterial substances to have had spatial locations?

    Consider other possible immaterial minds - such as ghosts and angels.  A ghost is thought of as haunting one particular house at a time.  But how could one explain a ghost's inability to haunt many houses at one and the same time?  The natural explanation, surely, is that it can affect material things only in locations that are within a certain distance of where the ghost itself is located.  If this explanation is rejected, it is not easy to see what explanation can be put in its place.  It seems very natural, then, to think of such immaterial substances as having spatial locations.

3.  Where Are Sense-Data Located?

 To defend his argument against sense-data, Mike needs to show that there is no satisfactory answer to the question of where sense-data are located.  He begins by suggesting that there are five non-arbitrary answers to the question of where sense-data are located:

“a. Sense data have no location.

b. Sense data are literally in your head.

c. Sense data are in the same places as the physical objects that cause them.  For instance, your sense datum of a table, caused by looking at a table, is located where that table is.

d. Sense data are located wherever they appear to be.

e. Sense data are located in an alternate space, separate from the space of physical objects.” (149)

4.  Answer 1.  “Sense Data Have No Location”

Mike Huemer's objection to this view is as follows:

"1.  In perception, I am immediately aware of things with spatial properties (things with shapes, sizes, and spatial relations to each other).

2.  Whatever has spatial properties has a location.

3.  Therefore, the things I am thus aware of have locations."   (150)

Comment

    It would seem that one can construct parallel arguments dealing with (1) bodily sensations,  (2) introspection of hallucinatory experiences, and (3) dreams, yielding the conclusion that bodily sensations, the contents of hallcinatory experiences - such as Macbeth's dagger - and the contents of dreams have location.  But if these conclusions are right, then the second premise of Mike Huemer's argument against sense-data - that is, that only physical things have loction - cannot be right.

5.  Answer 2. “Sense Data Are in Your Head”

Mike Huemer's basic argument against the view that sense-data are located in the head is contained in the following passage:

"Second: if visual sense data have the properties that I appear to see (e.g., "green", "rectangular"), what about tactile properties?  Shouldn't we also say that when I touch something I have tactile sense data and that these tactile sense data have the properties that I appear to be feeling?  I think the sense data theorist will agree.  Now, when I touch the book, I seem to feel something solid ('solid' here being used in the sense of 'hard', rather than 'filled all the way through')-in sense datum language, we would say I am having a sense datum of solidity.  Thus, my sense datum is actually solid.  So there is a solid, rectangular object inside my head.  It appears when I touch the book and disappears when I stop touching it.  Why doesn't it get in the way of the brain material that is actually there?  The old maxim that two things cannot occupy the same space at the same time would seem to apply here.  It is true that my brain is a bit mushy, so a solid object, if it wasn't too big, could push the brain material aside-but I doubt anyone believes that is what is happening."   (153)

 Comment

     Just as instances of qualitative greenness are not properties of external objects, so instances of qualitative, felt solidity are not properties of external objects.  The former are correlated with reflectance properties of external objects, and the latter with the ability of external objects to resist forces exerted on them.  It is the latter property that is the property of being solid.  The inference that Mike makes in the above passage in moving from "I am having a sense datum of solidity" to "Thus, my sense datum is actually solid" is, accordingly, unsound: property instances of qualitative felt solidity do not have an ability to resist forces exerted upon them, and so are not themselves solid.

6.  Two Arguments for the View that Sense-Data Are Located in the Head

6.1 The Causal Continuity Argument

     This first argument can be put as follows:

(1) Though it is not a necessary truth that causal connections and processes exhibit either spatial continuity or temporal continuity, in the actual world all causal connections and processes that have been closely studied turn out to be characterized by spatial continuity, in the sense that for any state of affairs that is a cause of some state of affairs that is not in the same location, one can find other states of affairs that are both spatially as close to the cause as one wants, and that are causally intermediate between the cause and the original effect.

(2) Therefore, it is likely that this is so for the causal connections from brain states as causes to mental states as effects.

(3) Therefore, since the cause is in the head, either the mental state must either be in the same location as the brain state that causes it, or it must be a temporally extended state, parts of which are spatially as close as one likes to the brain state, or else there must be intervening states that are neither brain states nor mental states.

(4) There is no reason for believing that there are causally intermediate states that are neither brain states nor mental states.

(5) Therefore, mental states must be either where their brain-state causes are, or else arbitrarily close to those brain-state causes.

 (6) Since the brain-state causes are inside the head, the mental states that they cause must also be inside the head.

6.2 The 'Form of the Relevant Causal Laws' Argument

    This argument, which was, in effect, alluded to earlier, can be put as follows:

(1) There must be causal laws connecting brain-states as causes to mental states as effects, and there are two main possibilities with regard to the form of such laws:

Hypothesis 1: For any location, S, the occurrence of a brain state of type B at location S causally gives rise to the occurrence of a mental state of type M at location S.

Hypothesis 2:  For any location, S, the occurrence of a brain state, X, of type B at location S causally gives rise to the occurrence of a mental state of type M in some mind P that stands in (the non-spatial) relation R to X

(2) Hypothesis 1 is simpler than Hypothesis 2, since the last postulates an extra relation R.

(3) Other thing being equal, the simpler hypothesis is to be preferred, since it is more likely to be true.

(4) Therefore, mental states are in the same location as the brain states that cause them.

(5) Therefore, mental states are in the head.

6.3 The Location of Sense-Data

    Both of the arguments just given are concerned with the location of mental states, rather than sense-data per se.  But one can now argue as follows:

(1) The instances of qualitative color properties that are involved in visual sense-data are property instances that are part of the relevant visual experience.

(2) Accordingly, those sense-data are located where the relevant experiences are.

(3) Experiences are mental states.

(4) Mental states have been shown to be located in the head.

(5) Therefore, experiences, in particular, are located in the head.

(6) Therefore, the sense-data involved in those experiences must also be located in the head.

7.  Alternative 3:  “Sense Data Are in the Same Place as the Distal Object”

Mike Huemer offers a number of objections to this alternative.  One important one is that when one applies this view in cases where one is seeing very distant objects, the result is incompatible with Einstein's Special Theory of Relativity.  One way of putting this objection is as follows:

(1) Given how sense-data are defined, they are part of an experience, and so exist at the same time as the relevant experience.

(2) If the sense-data exist at the time when one is having the relevant experiences, and if they also exist where the star is, then the causal process by which brain states give rise to experiences involving sense-data must traverse a distance of many light-years in a very brief period of time.

(3) Therefore, if the sense-data exist where the star is, the relevant causal process must traverse a distance of a thousand light-years in a very brief period of time.

(4) Such a causal process would be faster than the speed of light.

(5) Therefore, if the sense-data exist where the star is, the relevant causal process must travel faster than the speed of light.

(6) If the Special Theory of Relativity is true, it is impossible for any causal process to travel faster than the speed of light.

(7) Therefore, if the sense-data exist where the star is, the Special Theory of Relativity is false.

A second important objection involves the case of dreams and hallucinations:

"Consider dreams and hallucinations.  In these cases, there are no distal objects, so no places for the sense data to be located, according to the present theory.  Yet the indirect realist would be hard pressed to deny that sense data exist in these cases.  After all, hallucinations are one of the central kinds of phenomena that sense data are supposed to explain."  (156-7)

8.  Alternative 4:  “Sense Data Are Wherever The Appear to Be”

    This view is Frank Jackson's view.  Mike's initial comment on this view is as follows:

"This answer is the most natural one, if you're going to grant spatial properties to sense-data at all.  If sense-data have the shapes, colors, and other properties that we seem to perceive, why not also the positions?"  (157)

Comment

     The issue is a controversial one, but I think that there are good reasons for treating distance differently from color and shape.  One reason is this.  Position involves distance from something.  To be directly acquainted with an instance of a relation, however, one must be directly acquainted with all of the entities that enter into that instance of the relation.  But one is not directly acquainted with the something that different sense-data are claimed to be different distances from.  Accordingly, since one is not directly acquainted with one of the relata, the distance to a sense-datum from that object cannot be an object of direct acquaintance.

*********************************************************************************

    Mike's main objections to this view concerning the location of sense-data focus upon the cases of dreams and hallucinations:

(1) One objection concerns the case where one has a dream about a non-existent place, since then it is impossible for the sense-data to be where they seem to be.  (Here it might be argued, however, that in a dream the sense-data appear to be in front of one, and that that is where they are, rather than in the imaginary place where you are dreaming that you are.

(2) Mike's next objection involves the case of a brain in a vat, and here the point is that while, when one is not a brain in a vat, one can describe a sense datum as being in front of one, since one then has experiences of one's body, this is not possible if one is a brain in a vat, since then one has no experience of a body:

"The essential problem here is that, in certain cases, there could be no such thing as the place where a sense datum appears to be, even when the alleged sense datum would have to be one with spatial properties.  Of course, to the brain in a vat, it appears that the unicorn is somewhere.  But there is no real place such that the unicorn appears to be in that place.  And so the answer, 'Sense data are in the places where they appear to be,' does not work."  (159)

9.  Alternative 5:  “Sense Data Are in Phenomenal Space”

    Mike Huemer describes this fifth view as follows:

"Our fifth and final form of indirect realism holds that sense data exist in an alternate space, separate from the space that physical objects occupy.  The laws (if any) that apply to this alternate space may differ from the laws of physical space.  For instance, there may be no problem with transmitting influences faster than the speed of light in the alternate space, the alternate space may be Euclidean and unaffected by gravitational fields, and so on.  We can call this alternate space 'phenomenal space' to distinguish it from physical space."  (159)

    Mike advances three objections against this view that sense-data are in phenomenal space:

Objection 1:  This view conflicts with the Special Theory of Relativity, because sense-data then stand in temporal relations, but not in spatial relations, to physical objects.

Objection 2:  There is a problem about the causal relations between physical states of affairs and sense-data, so conceived, since one has causal relations between things that are not spatially related.

Objection 3:  There is a problem concerning the location of sense-data in phenomenal space, since there are no spatial relations between sense-data and physical things.

10.  Summing Up:  Spatial Relations, Locations, and Sense-Data

    What has emerged is that two distinctions are crucial for a satisfactory approach to questions concerning sense-data, spatial relations, and locations.

(1) The first distinction is that between (a) the locations of sense-data, and (b) spatial relations within and between sense-data.

(2) The second distinction is that between (a) purely relational spaces, and (b) substantival spaces.

When these distinctions are in place, one can see that the five alternatives that Mike Huemer considers do not exhaust the range of alternatives.  In particular, there is the following alternative:

Alternative 6:  The Existence of Sense-Data Involves Two Different Spaces

1.  All mental states, including experiences are located in the spacetime of physics.

2.  The instances of qualitative color properties with which visual sense-data are identical are part of visual experiences.

3.  Therefore, sense-data have locations in substantival, physical space.

4.  They are located where their immediate causes are located.

5.  Since their immediate causes are brain states, sense-data are located in the head.

6.  But instances of qualitative color properties have shapes, and stand in phenomenal spatial relations to other instances of qualitative color properties.

7.  Therefore, sense-data are also in a phenomenal space.

8.  But that space is purely relational, not substantival: it is not possible for such a space to exist in the absence of all instances of qualitative colors.

    All of the claims involved in this final alternative are, I think, plausible, and not open to any of the objections that Mike advances against the five alternatives that he considers.