I am worried this answer will crater the discussion that continues on the previous thread, so I’m opening a new thread to discuss Dennett’s lecture on his book Freedom Evolves. I know it is sort of stupid to post about this when I haven’t read the book yet, but oh well. Consider this a post about the lecture.
Daniel Dennett – Freedom Evolves – a Dangerous Idea Part 1 (9:55)
Daniel Dennett – Freedom Evolves – a Dangerous Idea Part 2 (9:58)
Daniel Dennett – Freedom Evolves – a Dangerous Idea Part 3 (10:01)
Daniel Dennett – Freedom Evolves – a Dangerous Idea Part 4 (9:59)
Daniel Dennett – Freedom Evolves – a Dangerous Idea Part 5 (9:48)
Daniel Dennett – Freedom Evolves – a Dangerous Idea Part 6 (3:01)
Dennett starts by concededing the obvious implications of determinism. Yes, a Laplacian Demon could calculate everything that will every happen, yes, there is only one future which is inevitable. However, he argues that once a person comes along who can see a brick coming, imagine the future of being hit by the brick, and duck so as to avoid being hit, this introduces an element of evitability to the universe which otherwise does not exist (inevitable=unavoidable; evitable=avoidable).
It is almost so obvious as to be insulting, but the use of inevitable and evitable in this analysis is entirely equivocal and appears to be intentionally designed to obfuscate. Based on his own analysis, the fact that he ducked to avoid the brick is inevitable and his being hit by the brick was evitable because of his inevitable duck. If there is no word game involved here may God strike me down.
So, let’s move on the what I think his real point is. He is demonstrating that from the standpoint of the person, getting hit by a brick was avoided and it was avoided by something that person did. Since we all experience the world from a perspective like that brick ducker, this shows that the things we do really have consequence in the world. Furthermore, because he used an example of a person imagining the future and ducking based on that imagination, he has provided an example which will show that people are free whereas other kinds of things (like rocks and machines) are not.
I would suggest that the evitability he is hinging his argument on is present in everything that has ever happened, so it is not particularly special. It is always true that a different thing could have happened if one participant in the event had behaved differently. These “other things that could have happened” are thus avoided at every turn. The fact that our parts in the causal chain represent real causes with consequence is maybe the most obvious bedrock position of compatibilism ever, so I understand it, but I don’t think it represents a contribution to the field.
The interesting part of his solution flows from his insistence that only certain kinds of events count as “avoidance.” Specifically, it is the events in which someone consciously tried to avoid a certain thing and then succeeded. This is crafty because it means only things like humans can be free, which is what we all want. It also leads to all of your demands about computational and counter-factual robustness. What I just said is perhaps slightly too restrictive, because he seems to be comfortable talking about chess playing programs as real “avoiders” of checkmate (and, thus free). I haven’t read his book (so please correct me if I’m wrong), but based on his presentation it appears that a thing qualifies as being free if it is purposeful. Something can be purposeful (1) if it was designed to serve a certain purpose (like a chess program) or (2) if it has a conscious desire and can purposefully act to realize that desire (like a person does).
Notice, however, that something need not be computational to fit that requirement. It could be a simple filter or sorter. Imagine a machine that has a single file line of bananas coming down it on a conveyor belt. It has an optical sensor that can gauge the color of the banana and if one of the bananas appears to brown, it swings its mechanical arm across so as to guide the brown banana to the banana bread factory on the right then back again to let the yellow and green bananas go to the left where they will be sold in Vons. This machine creates “evitability” in the same way as the person ducking to miss a brick. Without this optical sensor and mechanical arm, it was inevitable that a brown banana would end up in the yellow/green banana bucket, but now that this machine evolved an optical sensor and a mechanical arm brown bananas in the wrong bucket can be avoided. Brown bananas in the wrong bucket is now “evitable.” This should qualify as a a machine with free will according to Dennett’s argument.
Now, let’s suppose that instead of an optical sensor and a mechanical arm, the machine is sorting walnets by size (small and large) and it has a long shoot with small holes in it. The walnuts roll along and by the end of the shoot all the small ones have fallen through holes into the bucket for small walnuts and the big ones drop out the end of the shoot into the bucket for big walnuts. This also seems to qualify as creating evitability and freedom according to Dennett’s criteria.
We can keep simplifying the machine and it seems that as long as the machine serves a certain purpose we can describe that purpose in terms of avoiding a certain outcome and the machine will qualify as being free. All of his talk about competence is meaningful only in the context of accomplishing a certain purpose. As far as I can tell from his lecture, having a purpose is the principle ingredient of freedom.
The interesting thing is to talk about what moral responsibility and accountability mean in a framework like the one Dennett suggests, but before we open that can of worms, where have I gone wrong so far in representing his argument for what constitutes freedom?