Sunday, May 29, 2011

Substance Dualism (Part Two): Criticisms of The Basic Argument

In the previous post, I summarised William Jaworski’s presentation of the basic argument for substance dualism. In this post, I’ll be covering Jaworski’s discussion of the criticisms of that basic argument. Remember, substance dualism is a time-honoured position within the philosophy of mind and it’s well worth knowing why some find it persuasive and others do not.

As we saw the last day, the basic argument for substance dualism relies on two premises. The first of those premises derives from the axioms of identity and is relatively uncontroversial. The second of those premises makes a modal claim. It says that we (i.e. minds/persons) can exist without bodies. This claim is controversial and is supported by two equally controversial arguments: (i) the conceivability-possibility argument; and (ii) the essential property argument. Both were discussed in part one. Now we turn to the criticisms of both those arguments.

1. What can we actually conceive?
Let’s focus on the conceivability-possibility argument first. As we know, this argument has the following form:

  • (2.1) If it is conceivable that x can exist without y, then it is possible that x can exist without y.
  • (2.2) It is conceivable that we could exist without bodies (including brains).
  • (2) Therefore, we can exist without bodies.

There are obviously two points of attack here: premise (2.1) or premise (2.2). We’ll take these in reverse order.

Premise (2.2), as pointed out in part one, can be supported by three examples. Each of these examples purports to demonstrate that we can conceive of our existence without a body. The first example asks us to imagine existence in the after life; the second example asks us to imagine watching our own bodies being annihilated; and the third example asks us to imagine we are victims of a Cartesian Demon.

In the initial presentation we assumed that these examples really did involve the conception of a mind existing without a body. This time round we need to be more critical. Have we really thought about each example in sufficient and vivid detail to be sure that we can conceive of our own existence without the presence of a body? This is a serious question: if our imaginings are superficial and vague, we may only be conceiving of the partial absence of a body not its complete absence.

Consider the annihilation example (see description in previous post) and ask yourself: can you really conceive of yourself watching your body being annihilated? How do you suppose you receive the necessary visual input? Can you really imagine receiving it without any eyes?

Buried amongst these rhetorical questions we have our first attack on (2.2):

  • (3) Our imagining of the possibilities might be too vague and superficial to count as true conceivability.

Substance dualists can respond by saying only clear and distinct conceivability counts. This is acceptable but in that case it’s much less obvious that examples of the sort presented in the last post pass the test of conceivability.

2. Is the Conceivable really Possible?
Let’s turn our attention to premise (2.1) now. This premise proposes that conceivability is a reliable guide to possibility. Is this correct? Can our imaginations really tell us what obtains in different possible worlds? Well, it all depends on the type of possibility we are concerned with. Conceivability might be a reliable guide to some forms of possibility and not to others.

Jaworski orients his discussion around three types of possibility:

Technological Possibility: P is technologically possible if and only if P is consistent with current technological constraints.
Physical Possibility: P is physically possible if and only if P is consistent with the laws of physics.
Metaphysical Possibility: P is metaphysically possible if and only if P is consistent with itself.

These three types of possibility are very different from one another. It is technologically impossible (right now) to cure cancer, but this does not mean that it is physically impossible to do so. And it is physically impossible for there to be a 1000kg uranium sphere, but this does not mean that such an object is metaphysically impossible. There’s nothing internally inconsistent about the concept: if the laws of physics had been different, such an object could exist.

Conceivability is likely to be a poor guide to technological and physical possibility. Both concepts depend heavily on the current state of our empirical knowledge. We can’t now conceive of the technology that will cure cancer (if we could, we would simply invent it) but we might in the future be able to conceive of it. Likewise, being able to conclusively say that something is physically impossible depends on knowing all the laws of physics. But every physicist knows there are gaps in the current laws that need to be filled. It is empirical discovery, not a priori thought experiments, that flesh out our understanding of technological and physical possibility.

What then of metaphysical possibility? This is the brand of possibility to which the substance dualist appeals. Can he or she say, definitively, that non-material existence is possible solely on the basis of a thought experiment?

There do appear to be occasions on which a priori methods tell us something about metaphysical possibility. For instance, the rules of logic tell us that it is impossible for a married man to also be an unmarried man. Such a statement is self-referentially inconsistent. In a similar vein, the rules of semantics tell us that it is impossible for a bachelor to be married. Such a statement is analytically true.

If the independency of mind and body were simply a logical or analytical truth, the substance dualist might be home and dry. But this need not be the case. For there is a third type of metaphysical possibility that does seem to depend on empirical evidence. The classic example is the identity relation between “water” and H2O. This is thought to be a metaphysically necessary relation, but one that was only established through empirical investigation of the referent of the term “water”. (see Twin Earth thought experiments for more). If the relation of mind and body is like that of water and H2O, then a priori methods are insufficient to prove what needs to be proved.

So we come to the crux of the problem for the substance dualist:

  • (4) While conceivability might be a reliable guide to some types of metaphysical possibility, it is not a reliable guide to all types. Furthermore, it might be that the mind-body relation is a type of metaphysical necessity that can only be established through empirical investigation.

3. The Essential Property Argument
We turn now to the criticisms of the essential property argument. As we recall, the argument looked like this:
  • (2.3) An object, event or state of affairs can exist without its accidental properties; it cannot exist without its essential properties.
  • (2.4) Having a body is a strictly accidental property of beings like us.
  • (2) Therefore, we can exist without bodies.

Premise (2.3) is unobjectionable; premise (2.4) is more dubious. Last time out, we saw that there were two methods for establishing what counted as an essential property and what counted as an accidental property: (i) the conceptual method and (ii) the empirical method. The former was used by the substance dualist and was the basis on which (2.4) was reached.

It is time to call that method into question. Continuing with the line of thought established above when looking at metaphysical possibility, we can argue that conceptual thought experiments are an insufficient test for whether a property is essential or accidental. The water example can be called to assist us in making this point: a purely conceptual analysis would not have revealed that the essence of water is H2O -- empirical investigation was needed for that. Discovering the essential properties of persons or minds might require similar empirical investigation.

This brings us to a second critique of the essential property argument. According to this critique, the argument begs the question against physicalism. The reasoning behind the critique is as follows. Most will agree (substance dualists included) that if something is physical in nature, then only physics can give us the final word on its properties (essential or otherwise). But then this implies that if persons are physical, then only physics can give us the final word on their essential properties. The essential property argument assumes that the conceptual method can determine the essential properties of persons. Thus, the argument must be assuming that persons are non-physical. Therefore, it begs the question.

There is one final criticism of the argument. This was originally noted by Descartes’ contemporary Antoine Arnauld. Arnauld pointed out that, even if conceivability is a guide to metaphysical possibility, we can often fail to initially recognise all the properties that an object possesses. For instance, before the advent of electromagnetic theory, if we were listing all the properties of an object, we would have missed out on most (if not all) of its electromagnetic properties. After the advent of electromagnetic theory, we would have had a greatly enriched understanding of the properties possessed by the object.

So following Descartes’s procedure might lead us to miss out on some of the essential properties possessed by an object. We might also fail to appreciate the necessary connections between some of those properties. Applying this to the mind-body problem, following Descartes might lead to an impoverished understanding of the essential properties of a person.

We can summarise these three criticisms as follows:

  • (5) The conceptual method is insufficient for establishing the essential properties of all objects.
  • (6) The reliance on the conceptual method is question-begging.
  • (7) Descartes’s procedure may lead us to miss out on some of the essential properties of persons.

The completed map for all the criticisms of the second premise of the substance dualist argument can now be drawn up.

4. Moving forward
To this point in the discussion we have been focused on the basic argument for substance dualism and rebuttals to it. In the next post, we’ll start to consider the reasons for thinking substance dualism is false.


  1. This comment has been removed by the author.

  2. A little and nasty typo :)
    "In a similar vein, the rules of semantics tell us that it is impossible for a bachelor to be a married"

  3. I don't see the problem with that one. A bachelor is an unmarried man, so it would be impossible for there to be a married bachelor.

  4. Well, sorry I guess the "a" is a typo.

  5. I'm not sure conceivability even demonstrates possibility for matters of logical or analytical truth. Imagine that I write out a complex mathematical equation that, I claim, solves to equal 7. It is conceivable that the equation does in fact equal 7. It is also conceivable that I am wrong and the equation equals 8. And yet only one of those answers (at most) is metaphysically possible, and we would need to rely on improving our knowledge in order to resolve which one was which.

    In fact, while both might be conceivable right now, if you were to master the relevant mathematics sufficiently to obtain an intuitive grasp of the math used, it might cease to be conceivable to you that the answer could be either 7 or 8. Just as I cannot conceive of 2 + 2 equaling 5, because I understand too much of the workings of the equation to even imagine an incorrect answer.

  6. Well, let's use a concrete example. Was it, at one time, conceivable that Fermat's last theorem was false? I guess it probably was. A survey of the relevant subgroup of mathematician's prior to the publication of Andrew Wiles's might have revealed some thinking it was false and some thinking it was true.

    But is that true conceivability? Maybe the proponent of conceivability would fall back on the "clear and distinct" condition. I don't know. I'm not an expert on modal epistemology (although there is someone who has been known to read this blog who is. Maybe they could chime in on this point?).

  7. One glaring error in the initial argument seems to be passing without remark. That arguments uses an assertion of uncertainty to support a conclusion of certainty.

    "(2) We can exist without bodies." a statement of uncertainty (since no-one is claiming to have evidence that a person has existed without a body).

    As such, it cannot support the following statement:

    "(3) Therefore, we cannot be bodies."

    At most, it can support the statement that we "may be something other than bodies", or "we cannot be certain that we are bodies".

  8. Nevermind, this is essentially the same as your point 4 above.

  9. I'm a dualist but I don't subscribe to the substance dualist argument that you put forward in your opening post. (I don't see much evidence for existence beyond bodies) Here's my argument in favor of dualism. Planning is clearly a feature of our world. Planning requires that you treat identical units differently. If you want to build a staircase you have to treat each identical block in a different fashion. Physical law cannot build a staircase, nor can it plan anything because the laws of physics treat identical units identically.

    In order to plan you must choose which units go where. If this choice is caused by something else then it's not a choice. If this something else, say, a physical law, causes the human action, then planning cannot result because, as I said, physical laws cannot plan, they must treat all identical units identically.

    If choice therefore is uncaused and it must be uncaused then it cannot be just atoms, choice must be a force like gravity, that is to say immaterial, but it has the ability to make decisions, have intentions, etc. Choice has power to move which atoms it wants and where. But choice is limited, for example, there are certain masses which are too heavy for it to move.

    You simply cannot reconcile physicalism with a very obvious feature of our universe: planning.

  10. Yes, I saw it. Sorry, I'm preoccupied with another debate this week so I'm unlikely to be able to offer any sort of substantive response for a while. Also, if you're really looking for someone to discuss this critically with (as opposed to just nodding in assent or something) I'm not sure that I'm the ideal person since I neither study nor specialise in philosophy of mind.

    Of the top of my head, I'd say I can't see any reason why planning wouldn't just involve a law-like algorithm performing operations on objects. E.g. for making a stairs you'd have an algorithm that stipulated "add step 1 at height 1, add step X at height X, keep going until height Y is reached" or something like that. It seems possible to program computers and robots to perform these operations without necessarily implicating an immaterial substance.

  11. Where would the algorithm be physically located in space? In can't be located in the DNA because DNA is just nucleotides, so you would still need an immaterial intelligence to interpret the algorithmic code. Second, let's say that the algorithm is not written in the DNA but is just immaterial like F = MA. F = MA after all is immaterial, right? Well, would all these algorithmic codes be written before the Big Bang or after? Either way, it's very hard to imagine all these immaterial codes being written. Moreover, who writes the algorithmic codes? God? Third, up till now, as far as I'm aware, algorithms can't be written for any thing other than discreet objects and symbols. Reality is continuous, not discreet.