Maverick Philosopher

. . . you can predict its move with supreme confidence. Nothing about the laws of physics forces this move, and nothing about the specific design of the computer forces this move. The move is forced by the overwhelmingly good reasons for making it and not any other move. (Emphasis in the original)

Let's think about this. Suppose the computer's king is in check and can evade check in only one way, by moving to square h8. So the move Kh8 is forced. Now it is true that it is not forced by the laws of physics. But it is forced by the laws of physics PLUS the initial conditions. A Laplacean demon or a special supercomputer monitoring the chess playing computer could predict (in the case of the demon) or 'predict' (in the case of the supercomputer) that the forced move would be made. The prediction would be made with "supreme confidence" if determinism is true, or something very close to it if some quantum indeterminacy enters the picture. Instead of 'forced' we can say 'caused' and then take causation as either detrministic or probabilistic.

Similarly, though it is true that nothing about the specific design of the computer forces the Kh8 move, the move is forced by the specific design PLUS the initial inputs. The design stance involves higher-level description, but the physical and the design stances both allow predictability "with supreme confidence" IF the initial conditions/inputs are taken into consideration.

Thus Dennett's argument is unsound. He is arguing as follows:

1. The move is not forced by the physics
2. The move is not forced by the design
3. The move is forced
ERGO
4. The move is forced by good reasons
ERGO
5. The computer has good reasons and is therefore a mind, an intentional system.

What I have just shown, however, is that (1) and (2) are false.

What Dennett is trying to show is that a computer running a chess program has reasons for as opposed to mere causes of its behavior. If he can show that, then he can show that mind is present in the computer. To make his case he exploits the standard philosophical distinction between reasons and causes and tries to show that since there are no causes there must be reasons. But as we have just seen, it does not work. There are causes both at the level of the physical stance and at the level of the design stance. The availability of causes removes the ground for imputing reasons.

Dennett was talking about his chess playing computer and a particular forced move that it made. The question arose as to what forced it. We heard him say that "The move is forced by the overwhelmingly good reasons [the computer has] for making it and not any other move." (KM 31, emphasis in original; full bibliographical data in the prerequisite post.) But I smell some flatulence in the air, a conflation to be precise, a conflation of 'is forced' with 'can be viewed as forced from the intentional stance.'

I expect not to be contradicted if I say that 'X is Y' is not the same as 'X can be viewed as Y.' For example, 'The asteroid is a point-mass' says something different from 'The asteroid can be viewed as a point-mass for purposes of calculating its orbit.'

There is no doubt that I can and often do assume the intentional stance when playing chess against my computer: I view it as if it had beliefs, desire and purposes, as if it were an agent acting for reasons. The physical stance and the design stance are too complicated when it comes to predicting the computer's behavior, so I impute to it beliefs, desires, purposes and reasons. I view it as wanting to checkmate me, as believing that a back rank mate is in the offing, etc.

So there is no doubt that a particular forced move can be viewed as forced by reasons that the computer has for making it. But that is not to say that the move is forced by reasons.

But it is precisely this latter claim that Dennett needs if he is to build his gradualist bridge from mindlessness to mind, in particular, from physical causes to reasons. He needs to be able to find intentionality and agency in computers, thermostats, and indeed in self-replicating macromolecules. (KM 34)

In brief, Dennett needs the is, but how does he get to it from the can be viewed as?

The whole conjuring trick seems to require a tripartite conflation of original (intrinsic), derived, and 'as if' intentionality. What Dennett needs to say and apparently is saying is that no intentionality is intrinsic or original, and no intentionality is merely 'as if.' All intentionality is genuine but derivative. And this derivative intentionality is apparently ubiquitous. Even a lock and key is an intentional system since the lock represents the key that opens it. "A lock and key exhibit the crudest form of intentionality; so do the opioid receptors in brain cells — receptors that are designed to accept the endorphin molecules that nature has been providing in brains for millions of years." (KM 35)

One obvious question is this: How could ALL intentionality be derivative? The intentionality of a picture I draw gets its intentionality from me, or my brain. Whence the intentionality of my brain states? To avoid a vicious infinite regress, must there not be some original intentionality somewhere?

Dennett's answer is that ". . . the brain is an artifact, and it gets whatever intentionality its parts have from their role in the ongoing economy of the larger system of which it is a part — or, in other words, from the intentions of its creator, Mother Nature (otherwise known as the process of evolution by natural selection)."

The idea, I take it, is that there is a sort of finite regress to ever smaller and dumber and cruder intentional systems. The intentionality of the smarter systems derives from that of the dumber systems until we arrive at systems so dumb that their intentionality derives from mindless matter.

But then how explain the transition from mindless matter to the most primitive form of intentionality? Or is Dennnett a closet panpsychist who holds that intentionality is ubiquitous?

It looks like a dilemma: either panpsychism or an inexplicable and mysterious jump from mindless matter to primitive intentionality. But if there must be a mysterious jump somewhere, why is Dennett's view preferable to dualism?

If you press down a wrinkle in a carpet in one place only to have it appear in another place, then you still have a wrinkle -- under a coffee table perhaps where it is less visible. Similarly, burying a mysterious transition deep in the evolutionary past may make it less obvious but does not remove it.

Thanks for the interesting and fruitful comments. There are properties that it makes sense to say are emergent. Consider the property of being a heap. A grain of sand is not a heap. But add 9,999 grains to it and one definitely has a heap. Somewhere between one grain and 10,000 grains the heap emerges, though one will be hard pressed to say exactly where, and why adding one more grain should suddenly make a difference. Perhaps you are saying that intentionality is like this. At a certain threshhold of complexity intentionality emerges.

But note that emergentism is a kind of dualism: what emerges (intentionality) is radically different from its emergence base, mindless matter. Since emergentism is a form of dualism, albeit not a form of substance dualism, it cannot be used to make sense of Dennett's position which eschews as unscientific every sort of dualism, not just substance dualism.

Note also that it cannot be Dennett's view that original (intrinsic) intentionality emerges. For he explicitly denies that there is an original intentionality. (For example, at the end of "Fast Thinking in The Intentonal Stance, p. 337: "There is no such thing as intrinsic intentionality . . . .")

This implies that all intentionality is derivative. The intentionality of smarter systems derives from that of dumber and dumber subsystems until we arrive at subsystems so dumb that they are utterly free of intentionality (like the solitary grain of sand that is 'heapless').

Intentionality is a useful model for understanding the design and behavior of a system, but I think we must be careful not to grant it some sort of mysterious soul-like status. Obviously a stone has no intentionality, while we have it in spades. But I think Dennett has done a good job of showing how intentionality could ramp into existence gradually, once matter has organized itself into primitive replicators. If a simple organism has over eons evolved cellular mechanisms for engulfing food, then we can read intentionality into those new mechanisms and structures (they are "for" eating), though nothing inexplicable or mysterious has happened.

You say that intentionality is a model, and then you say it can "ramp into existence." There is a tension here. If something comes into existence then it is real and not a mere model of the real. Are you perhaps conflating the intentional stance with intentionality?

Suppose we "read intentionality" into some cellular mechanism. No doubt we can do this just as I can ascribe beliefs and desires to my chess playing computer. But it doesn't follow that there is any intentionality in the cellular mechanism or in the computer. Note also that ascribing is itself a case of intentionality. So if there is no intrinsic intentionality, then my ascribing is intentional in virtue of some other entity's ascribing intentionality to me -- which appears to ignite a vicious infinite regress.

In Dennett's view even the mental mechanisms that underlie our consciousness (with all its rich and multifaceted intentionality) are just a logical extension of that process, a synergistic result of the joint efforts of simpler mental modules, which in turn are the result of the actions of bundles of neurons, and so forth. Our fabulously rich high-level intentionality emerges from all of this in the same way that "performance" or "handling" emerges from the parts of a Porsche. Yes, the individual car parts were designed by the intentionality of German engineers — but our mental parts were also designed by natural selection over deep time.

There is no doubt that you understand what Dennett WANTS to say, but the question remains as to whether or not what he wants to say can be coherently said.

Dennett's theory is ascriptivist: nothing has intentionality intrinsically; intentionality is ascribed or imputed or projected into a system from without. But this ascribing is itself an instance of intentionality, and you want to say that this ascribing is the "synergistic result of simpler mental modules" which are in turn the result of even simpler modules which are "designed by natural selection."

There is a deep unclarity here. How can D's theory be both ascriptivist and emergentist? You said a while back that we ascribe intentionality to primitive cellular mechanisms. Fine. But that implies that those organisms have no intrinsic intentionality. How then can enough of them working together -- syn-ergy -- give rise to the intentionality that ascribes intentionality to them?

There is a vicious circle here. X ascribes intentionality to a, b, and c. Apart from this ascription they would exhibit no intentionality. But X's ascribing is an instance of intentionality that results from the synergy of a, b, and c -- which presupposes that there is some primitive intentionality in a, b, and c that is not the result of X's ascription.

There is also the unclarity of "designed by natural selection." Strictly speaking, natural selection doesn't design anything. Designing is an instance of intentionality, and intentionality emerges only way down the evolutionary road. What Dennett does is help himself to, or presuppose, the very thing he is trying to account for, namely, intentionality.

As for a bridge from causes to reasons, why NOT say that the computer's move is forced by reasons? Yes, we can take an extreme-reductionist view and try to determine the computer's move from the state of its constituent particles, just as we might for a human player, but why bother? We know very well that the computer will avoid mate in one because if it doesn't it will lose , just the same as a good human player. Take any decent chess computer (and there are lots and lots of them - I have about seven or eight myself), and set up an avoidable mate-in-one, and the machine will always avoid it, regardless of the particulars of the hardware or software . I think this is amply sufficient to move from "can be viewed as" to "is". Yes, a human playing a child might lose on purpose (I often did as I was teaching my son the game), but that of course is because a human is a vastly more complex intentional system.

You can impute reasons to a computer or a thermostat or a circuit breaker: it has a reason to trip: it doesn't want the house to burn down and it with it. You could say it senses a surge in amperage caused by a short circuit. But of course it senses nothing: it is just a crude mechanical device. So I fail to see how viewing it as having reasons suffices to show that it has reasons for its 'actions' which of course are not actions strictly speaking.

I think your argument that there is something important left to be explained depends entirely on reifying intentionality, quite unnecessarily, into something more exotic than it needs to be.

I plead innocent. To reify is to make into something real that which is unreal. But intrinsic intentionality is real, and is indeed presupposed as real by any affirmation of derived intentionality. What you and Dennett are doing is denying a plain fact because it fails to fit into your preconceived naturalistic schema. And the cost of that denial is all sorts of absurdities: vicious infinite regresses, vicious circles, equivocations, etc.