Blog Coordinator

« Automatizing Mid-level Decision-making | Main | Nahmias on Free Will at Scientific American »

12/29/2014

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Peter,
I agree that imagination is central. Thanks for spelling out some of the neuropsychological details. All our philosophical psychology improves when scientifically informed. It's been a great month.

Another great post! Peter you have obviously worried a great long time about all these matters, and with Paul I once again pay respect to your indefatigable ability to defend your views in the most sweeping empirically- and philosophically-informed ways.

Would it be fair to say that the neurophysiological nature of human WMRs (i) add to non-human aspects of WMRs by utilizing indeterministic biophysics to increase associations (maybe symbolic) that are "recognizable by the executive mechanisms" (if that's a right phrase of your view) of the brain, and (ii) as so recognized contain mental contents that might be otherwise stated as propositions that are logically possible? In other words the power of the human brain is to entertain logically possible propositions and rate them as more or less relevant to practically entertained circumstances and other (perhaps less practical) goals? What I mean is that I can think about going through a doorway, and logically it is possible that I not only go right or left into a hallway, but sink ethereally through the floor or be beamed up to the Enterprise, but rate the latter two as impractical to other held future goals of getting to my next class. But if I'm writing a sci-fi story, then beaming up might be germane to achieving my narrative goals in a way that turning left in the hall may not. If that's what you're saying, that all makes good sense.

My problem is the fact that while we can be self-reflective about logical possibilities, that is not necessarily indicative of any particular metaphysical role those propositions as contents might play in the underlying process. I guess this all gets back to Kim-like worries. I can imagine that I am some sort of supercomputing bio-wetware that has evolved to do the following: (1) produce a self-observed display of logical possibilities that are relevant to some other like-displayed process of a goal-oriented sort to which the logical-possibilities display is recognized as relevant (I guess this is a crude way to describe reflective self-consciousness) and yet (2) throughout be running at the metaphysical level a program--perhaps even a program with indeterministic elements that play only a random role in the outcome--that I am completely unaware of (of course this must be the case) but of which I am also only passively the constitutive sum. Perhaps even the sense of controlled agency is part of the program; maybe I do indeed feel the sense of choosing and the onus of choice. All this seems like an account that is logically possible (and I'm certainly conceiving it no matter what is the case!), and yet dissolves any real basis for agential control of the LFW type.

I guess I'll express this in a very basic way: how do we get from the fact that we can conceive of logical possibilities to a firm foundation that our ability to do that plays a significant metaphysical role in agency that eliminates my worry above?

Alan, I spent the morning on today's post, and now have to drive back to NH from NYC. I will reply to you tonight. Sorry for the delay.

Alan,

I am slowly getting to answering comments that I have so far not been able to answer because of the many family holiday obligations this past week.

“Peter you have obviously worried a great long time about all these matters, and with Paul I once again pay respect to your indefatigable ability to defend your views in the most sweeping empirically- and philosophically-informed ways.”

Thanks for your kind words. Yes, I have worried about what makes our brains/minds so different from those of other animals, including a chimpanzee, for a long time. I have also wondered about the roots of human evil and good since I was young. I am half German and half Chinese, and saw instances of evil and good on both sides, in the history of my extended family in both Germany and China. For example, I loved my German grandfather very much. But as I became morally aware, I asked increasingly about the war. I discovered that he and his brother had been officers in the nazi party, that he had had many Zwangsarbeiter (slaves) at the factory in Cologne where he was head engineer in the design and construction of train wagons. I learned many things that seemed so incongruent with the funny and loving old opa I knew. I knew he was not an evil man, but a kind man, and yet, he and so many other Germans had committed so many acts of evil and metaevil. How could this be? This was probably the beginning of my questions about volition and free will. How could a kind man have freely chosen to do what he did? Later in life I read Hannah Arendt, learned about the Nuremberg trials, and the ‘banality of evil’ and the notion that ‘just following orders’ was no justification for decisions that are immoral. I think real life decision-making is not so simple, black-or-white, or easily condemned as just plain personal evil. The power of institutions in our lives can make it very difficult to act in a way that we would want to if we were following our own internal moral compass. When I talked with my opa about this, he said he had two daughters and a wife to feed and was not involved in the camps or the war machine, just in making train parts, so minded his own business, made money, and played his small role as a cog in a much larger machine that he felt no responsibility for. He recounted how desperate the times were during the Weimar hyperinflation and afterwards, how he had had no work, and how he had voted for Hitler in 1933 because he wanted work.

These and many other incongruencies have made me realize that the problem we face now is not the problem of human evil (though evil is of course a perennial problem), but the problem of human metaevil. We humans have a great task ahead of us, which is to figure out how to control artificial decision-making systems, such as corporations and governments, such that they cannot become metapsychopathic, tyrannical or totalitarian, and reinvent them in ways that enhance rather than harm human life. The advent of the information age has increased the problem of informational control by institutions manyfold. Computerization of our lives has greatly enhanced the potential for metaevil. Just think that the NSA might have the power to listen to and watch you through your laptop right now. The Stasi would have loved that. Don't believe me, listen to NSA whistleblower William Binney:

http://www.youtube.com/watch?v=TuET0kpHoyM

Institutional metaevil poses as great a moral challenge to humanity as slavery did in an earlier age, in my opinion. Moreover, our civilization is facing its biggest challenge since the rise of fascism in the 20s/30s/40s. In fact there are many similarities between what is happening now and the rise of fascism.

The problem and the consequences are dire, but the set of ideas to address the challenge we face as a civilization have not yet come into focus. Political leaders like Teddy Roosevelt or Andrew Jackson or Thomas Jefferson have yet to emerge on the national stage. Thinkers like Voltaire or Montesquieu have yet to fully shape the ideas that will reshape the world. This will all unfold this century.

In the short term I fear we are lurching toward a proxy war in the Ukraine this year. The neoconservatives who dominate the deep state realize that control of Eurasia is central to American hegemony, and that Ukraine is key to containing the rising power of Russia. As a dark Russian joke goes, the neocons are prepared to fight Russia down to the last Ukrainean. Here is an eye-opening 1997 quote from deep-state thinker Zbigniew Brzezinski:

“Without Ukraine Russia ceases to be empire, while with Ukraine - bought off first and subdued afterwards, it automatically turns into empire…(...) the new world order under the hegemony of the United States is created against Russia and on the fragments of Russia. Ukraine is the Western outpost to prevent the recreation of the Soviet Union.”

But the neocons in the State Dept. have played their cards incredibly clumsily, installed an unsavory nazi-istic puppet regime in Kiev, lost Crimea, and have triggered an alliance between Russia and China that is a genuine threat to the longterm hegemony of the US. And the likely Russian-Chinese plan to de-dollarize the world and perhaps create a new currency backed by gold/oil is a very direct threat to the American empire built on the petrodollar reserve currency and high-tech military-intelligence system. I honestly fear that so much is at stake for the neocon deep state, that it will trigger WW3 in an attempt to maintain control. I fear very dark times are ahead for humanity in coming decades.

There is a need for intellectuals to speak up and play their traditional role in shaping the ideas that can drive human actions. I hope philosophers and other intellectuals are up to addressing the serious challenges our civilization now faces.

Alan,

“Would it be fair to say that the neurophysiological nature of human WMRs (i) add to non-human aspects of WMRs by utilizing indeterministic biophysics to increase associations (maybe symbolic) that are "recognizable by the executive mechanisms" (if that's a right phrase of your view) of the brain, and (ii) as so recognized contain mental contents that might be otherwise stated as propositions that are logically possible?”

I do not think that thought needs to be expressed as propositions. The kind of mental representations and operations we have been looking at in my lab are visual and perhaps visuomotoric. If you picture a chair and make it rotate in your mind, is that a propositional format? It would seem to me to be a sensory or sensorimotoric format. When you dream, do you dream in propositions or in images?

”In other words the power of the human brain is to entertain logically possible propositions and rate them as more or less relevant to practically entertained circumstances and other (perhaps less practical) goals?”

I think brain circuits can generate and evaluate propositions. I can evaluate whether it is true or not that I can walk through the wall. But I do not think imaginings have to be expressable as (logically possible) propositions. Propositions are symbolic in nature, but I would think that the non-symbolic brain of a dog is capable of some very complex thoughts in a non-propositional format. If someone imagines a human body with the head of a bee, that might have nothing to do with reality, or with what is logically possible in reality. It might just be creativity or art for art’s sake. So I am skeptical that thought is inherently propositional. An emotion of mirth or anger, for example, would seem to me to be entirely non-propositional. A mood can arise for no apparent reason at all, and need not be about anything, as a proposition would seem to always be.

In fact, every night, when we dream, we generate some pretty far out stuff. Much of it is not logical or possible. I find dreaming fascinating and do not think neuroscience understands it too well. The dominant view seems to be that of Alan Hobson, which is that dreams are effectively ‘interpreted randomness.’ That is, random inputs come in and higher level areas try to make sense of it. But a repeating dream cannot be accounted for in terms of randomness, because it is highly unlikely that a random sequence would happen again and again. So I doubt Hobson’s view is the last word on the matter of why the contents of dreams are what they are. At the other extreme, the ideas of Freud or Jung seem too mythological to easily fit into a modern neuroscientific view. By the way, is there any philosophical work on dreams or why the content of dreams is what it is?

“What I mean is that I can think about going through a doorway, and logically it is possible that I not only go right or left into a hallway, but sink ethereally through the floor or be beamed up to the Enterprise, but rate the latter two as impractical to other held future goals of getting to my next class. But if I'm writing a sci-fi story, then beaming up might be germane to achieving my narrative goals in a way that turning left in the hall may not.”

That sounds reasonable. I would say that constraints are also placed on imagination before possibilities are generated, so that only useful possibilities will be generated for conscious consideration or evaluation. For example, let’s say I have just dropped my car keys down the sewer. I can say ‘OK bag of tricks, please come up with some tool that might be in my environment that will allow me to poke through the grating to fish out my keys.’ I might then think of a wire coat hanger, evaluate it, modify its shape, and then use that.

“My problem is the fact that while we can be self-reflective about logical possibilities, that is not necessarily indicative of any particular metaphysical role those propositions as contents might play in the underlying process. I guess this all gets back to Kim-like worries. I can imagine that I am some sort of supercomputing bio-wetware that has evolved to do the following: (1) produce a self-observed display of logical possibilities that are relevant to some other like-displayed process of a goal-oriented sort to which the logical-possibilities display is recognized as relevant (I guess this is a crude way to describe reflective self-consciousness) and yet (2) throughout be running at the metaphysical level a program--perhaps even a program with indeterministic elements that play only a random role in the outcome--that I am completely unaware of (of course this must be the case) but of which I am also only passively the constitutive sum. Perhaps even the sense of controlled agency is part of the program; maybe I do indeed feel the sense of choosing and the onus of choice. All this seems like an account that is logically possible (and I'm certainly conceiving it no matter what is the case!), and yet dissolves any real basis for agential control of the LFW type. I guess I'll express this in a very basic way: how do we get from the fact that we can conceive of logical possibilities to a firm foundation that our ability to do that plays a significant metaphysical role in agency that eliminates my worry above?”

I am not sure I fully understood what you wrote above Alan. Is your point that mental events, such as the possibilities we consider, might not be causal? Well, certainly mental events are not free-floating entities. They have to be realized in some physical basis, just like a cup has to be realized in some physical basis. A cup is defined by its shape and its function of holding liquid, but a cup shape or a cup function cannot exist independently of some material realization of that shape and function. That said, it is multiply realizable in countless materials. Then you can turn around and say, well, that function of holding water had nothing to do with cupness per say, it does not hold water *qua* cupness, but is simply a consequence of local material interactions. But that also cannot be right, because if there is a hole or crack anywhere in the cup, it will not hold liquid. So the function emerges because of a global shape and set of global physical relationships, and cupness is just those relationships or that pattern. Similarly, an airplane wing will only function as a wing if there is a certain global shape. The lift afforded by Bernoulli’s principle emerges because air has to traverse a longer distance on top of than below the wing. Lift is not caused by any local portion of a wing. It is caused by the global interaction of air with the global shape of the wing. The brain is even more interesting than these examples of non-localistic causal emergence because detectors have evolved to respond to global patterns of neural input that only arise when certain informational facts are the case (say a ‘face neuron’ will only fire when a face is really present in the sensory input). And what global patterns a neuron is tuned to can be changed by changing its synaptic weights. A neuron that was tuned to this might come to be tuned to that.

Information realized in neural events is causal in virtue of placing constraints on allowable future neural activity. This is physically realized *informational* causation, because only by understanding what the informational constraints are will you understand why a neuron will fire. Two physical inputs might be extremely similar physically, but in one case a neuron will fire, and in another it will not, because the slight physical difference led to a difference of informational content or the meeting of informational criteria versus not. Conversely, two radically different physical patterns of input might count as the same informational input. Let me give a concrete example. My brain has been wired up to distinguish the sounds ‘ra’ and ‘la’ because I grew up in an English-speaking environment. I can hear very different inputs (say darth vader’s voice and mickey mouse’s voice) and will classify both as ‘ra’, say, if both darth vader and mickey mouse say ‘ram.’ Conversely, I can hear very similar inputs, say mickey mouse says ‘ram’ and ‘lamb,’ and then classify these as different. But this difference emerges because the parameters in place in assessing inputs are such that a boundary is placed between these phonemes. So, given how my brain is parameterized, the difference between ‘ra’ and ‘la’ is informative and makes a difference in how neural activity will respond to inputs. But a Japanese person’s brain cannot distinguish these phonemes. Under their parameters for processing input, ‘ram’ and ‘lamb’ get classified as identical information. Thus, as I have said before, information is fundamentally about acts of decoding. Even though such acts of decoding are physically realized, the only way to understand what will happen next is to analyze the system at the level of informational parameterization and informational decoding. At that level, physical properties like mass and momentum and other explananda that are needed to understand and explain physical causation at the level of particles, are not needed or useful. I stop at a stop sign, not because some force makes me stop, but because an informational rule tells me to stop. Informational is causal in the universe because informational parameters only permit possible physical causal chains that are also informational causal chains to happen. To understand which informational causal chain happened, you have to analyze the chain of causation at a level of information causing information, even though it is always physically realized.

If I have failed to answer or understand your worry, please try stating it again for me.

The comments to this entry are closed.

Books about Agency


3QD Prize 2014: Marcus Arvan