eBook Paul M. Churchland À À The Engine of Reason The Seat of the Soul A Philosophical

A picture of the mind is emerging and explanations now exist for what has so long seemed mysterious Philosopher Paul Churchland explains these scientific developments in a simple authoritative fashion He not only opens the door into the ongoing research of the neurobiological and connectionist communities but goes further probing the social and moral dimensions of recent experimental results that assign consciousness to all but the very simplest forms of animals


10 thoughts on “The Engine of Reason The Seat of the Soul A Philosophical Journey into the Brain

  1. says:

    In calculus one can find the area under any curve by filling it with rectangles and then making the rectangles smaller and smaller until there is no space left unfilled In neurophilosophy one can explain any phenomenon of consciousness by describing neurological functioning and then making the description and detailed until there is nothing left to explainThis is the agenda of Paul Churchland and neurophilosophy and I am not sure if he succeeds because there is nothing left to explain or because he has worn me out with the exercise His descriptions are beautifully complicated and they are not just inventive or ingenious but rather thoroughly methodical and soundly scientific The uestion is whether they really address the philosophical problemsMuch of what he does seems to be mathematical modeling which would be theoretical neurology than descriptive or empirical neurology He starts with the model of parallel distributed processing PDP such as a computer might do although most computers work on the basis of serial processing In PDP one starts with a variety of digitalized input which could be sensory and through a process of self correction and refinement one arrives at digital output which is the completion of a cognitive task For example one can take input from the optic nerves run it through PDP and come out with “Yes a suare” Obviously I am glossing over many detailsBut the techniue is strong enough to account for progressively complicated cognitive tasks recognition of shapes letters colors tastes sounds words faces emotional expressions grammatical patterns social situations in time etc etc After a point one wonders if there really is anything left to the life of the mind that cannot be accounted for in this wayChurchland identifies the three knottiest problems of reducing consciousness to neural processing as 1 meaning or semantic content Can PDP do than manipulate symbols? 2 mathematics Can PDP do math as humans do including Kurt Godel? and 3 the old problem of sense ualia Can PDP see red? He makes an energetic charge on each of these problems discussing the research with PDP which would suggest progress in addressing them He himself acknowledges no final conclusions on these uestions but the efforts being marshaled to answer them are formidableThis book feels like the textbook for a new science a new paradigm for neurology The logical development is exciting In the past when I studied neurology it has been like studying a map of Afghanistan But in this book I see a techniue for using PDP laid out with its conseuences which is similar to other epiphanies in science such as encountering Euclid’s geometry or chemical bonds or plate tectonics for the first time I am grateful for the experience Yet on a gut level I resist the effort to reduce or eliminate the mind even in a theoretical sense Churchland himself occasionally speaks of the self or the soul in an informal sense but as distinct from the mind I can’t help but feel that it is with the mind that we understand the universe including neurology It is the mind that finds the orderliness in chaos and sums it up in general principles How then can the mind explain itself away like the Cheshire cat disappearing and leaving only its smile? How can one use wonderful resources to explain away those wonderful resources? One can perhaps explain everything except the explainer


  2. says:

    Churchland's Engine of Reason is dated and that's too bad because I actually think it is one of his well written pieces probably even than Plato's Camera which I gave 5 stars The problem is that the change of material in neuroscience and cognitive science make it hard for writing to stay current for very long now; that's not Churchland's fault but the way that he structures the book makes it far easier for the content to become obsolete compared to contemporaries like Dan DennettI think that what should be said is that one of the reasons that Engine of Reason has become dated is part of what makes it great as a piece of philosophy it is current and in touch with the scientific literature that surrounds it That isn't to say that other philosophers aren't but many are hesitant to bombard readers with scientific content Churchland is unabashed about his knowledge of the contemporary literature in the science and as a result the book feels almost like something that would be written for an interdisciplinary audience attempting to talk about the neurosciences than anything in philosophy This is one thing that I absolutely love about ChurchlandThe major criticism that will get tossed at Churchland will come down that same line This doesn't feel like philosophy That's true Of course if you some of the content of the book you start to understand why This book is a part of a conversation in philosophy of mind that happened around the time of the publishing where philosophers raised concerns about the invocation of traditional intuitive concepts that seemed totally inviable for moving the discussion forward because the concepts themselves were either incoherent not meaningful or not supported by the science Churchland was largely responsible for championing the view that the terms that don't seem to be doing work should take a backseat to the things that can That is a fight that Churchland seems to have won in the technical literaturePaul Churchland's influence on contemporary philosophy often as the far wing of a spectrum of people concerned with the ability to talk about consciousness in strictly neural terms is hard to minimize He has had a major impact as a philosopher of science generally and is one of the best ways into a scientifically literate philosophy of mind Churchland's attempt to understand a broad array of topics in the field is something that should be admired and his attempt to bring that technical knowledge into the conversation is something that he shares only with the best in professional philosophy of mindThere are some issues that I take with the literature that don't seem to have gone away from Churchland's writing He's always very neurocentric He is incredibly interested in the metaphor of digital computation while acknowledging that the brain is clearly an analogue entity He limits himself a lot through the use of that approach which has become something of a standard for himI do think that for those looking to seriously pursue philosophy of mind this is a must read at some point largely because it fleshes out some of his points about the use of vectors and why they matter However I think that Plato's Camera is something to be taken a little heavily as it is far developed in terms of the ideas in philosophy of mind Also this really isn't for those just looking for a casual introduction; make sure you have some familiarity with the literature before diving into any of Paul Churchland's work


  3. says:

    This book sets forth Paul M Churchland’s program for understanding the mind by understanding the brain The approach that he discusses is parallel distributed processing PDP also known as neural nets The basics behind PDPs are networks that consists of three layers an input layer a middle layer and an output layer The layers are massively connected with than one unit being connected to other units The input layer receives some form of data like sensory information The middle layer is than computed using random values at the start but than these are adjust depending on how close the output layer matches the input layer This is called training the network Once the network has had sufficient training it is able to correctly output the appropriate interpretation of the input information a large percentage of the timeThe reason behind using these networks to process information is it is thought to be similar to brain networks of nerve cells massively connect to each other Of course the brain’s neurons are vastly connected than the simple networks researchers in this field use to explore the supposed workings of the brain But the sense in this field of computation is that while these networks setup on a computer are simpler than the brain’s connections of brain cells it captures the way the brain actually worksThere are some pretty impressive results using PDPs Churchland gives examples throughout the first part of the book including facial recognition stereoscopic vision 3 D vision sonar perception vocal production sensorimotor coordination recognition of facial emotional expressions and grammar production In addition in part one to the presentation of these various networks Churchland provides analysis in the form of vector coding which provides areas of prototypical zones The final chapter to this first part of the book is the use of neuroimaging and various brain disorders which are the results of damage to certain brain networks such as language processing facial recognition memory problems and mental illnessIn the second part of the book Churchland moves into the philosophical sphere The first thing he tackles is “The Puzzle of Consciousness” Here he presents a list of seven things that need explaining He also goes over some of the negative views on being able to explain consciousness and goes on to refute them He also posits what might be happening in the brain while we are conscious Next up is machine consciousness and the chances that this could happen After this he dives in to cultural aspects dealing with language science politics and art Finally he covers the merger of humans and technology what that might entail and the view of ourselves that might occurThe first uestion I have about PDPs is if they work like the network of neurons in the brain what works in the brain to provide the corrections seen in the PDPs? Also these PDPs might not even be implemented in the brain and Churchland does not provide any evidence beyond the structural that the brain actually works like the PDPs he describes I find support for my notion that we do not think in language when Churchland writes “In fact the cognitive priority of the preverbal over the verbal shows itself upon examination to be feature of almost all our cognitive categories” page 144 Again “ cognitive activities are exceedingly unlikely to have such a sentence like and inference like structure the basic unit of animal and human cognition is not the sententially expressible state ” page 182I will say that when Churchland says “it is a logically possible neurocomputational account of the phenomenon of consciousness” page 223 italics in original I think yes if the brain uses neurocomputation as describe in the PDPs presented in the book Churchland discusses and critiues Roger Penrose’s argument that a machine could not do mathematics I would add that a machine could be programmed to add the Godel statement it might run up against as a new axiom to the formal system under consideration This weakens Penrose’s argument in The Emperor's New Mind that brains do not work like computers not that I think that only for different reasons Another thing I would point out is that formal systems are not the everyday practice of mathematicians so machines could very well do mathematical proofsHere are a few thoughts about what Churchland writes in the book I wondered if it was necessary for Churchland to defend Thomas Kuhn’s philosophy of science in order to show that paradigms are important in his philosophy of mind Also when Churchland argues for prototypes and not rules as determining the actions of neural networks I thought couldn’t the prototypes enact the rules others think are indicative of learningFinally While Churchland does not mention eliminative materialism which is critical of folk psychology by name it is evident I still think eliminative materialism carries a huge promissory note to it Also I uestion why neural nets cannot produce the contents of folk psychology such as beliefs desires intentions etc as prototypes in vector space We surely have these contents in our mental life so according to him they should be found in the prototypes in vector spaces of the computations of PDPs While not uite a great book I did find it a very good one The main reason for not placing it in the first category is Churchland’s presentation of the PDPs especially those on perception which is not a major thing of interest to me in both brain science and the philosophy of mind It was certainly a thought provoking book and while I have my uestions about the actual reality of his premises they do appear to possibly be a workable solution in the end Admirably he is trying to understand how the brain produces the mind and not ignoring what neuroscience showsThis book is not for the philosophical computational and brain science faint of heart But if you want to dive into these areas of concern you will probably find it both profitable for thought and enjoyable to bootI will say one final thing I am glad I have such a wondrous thing as a brain


  4. says:

    The title of the book The Engine of Reason the Seat of the Soul provides promise Churchland's subtitle A Philosophical Journey into the Soul entices further But the first two thirds of his book is heavily technical and difficult His philosophical journey in the latter part of the book is only marginally easier to digest Churchland's intent to tie our cognition to others in the animal kingdom is good Also good is his argument that we sense the world in terms of multi sensory pictures gestalts? which he states is fundamentally different and far richer than can be expressed through serial seuential processing language For example morality based rules and principles limit the way we experience good and bad in the world On the subject of morality Churchland dismisses the moral skeptic's uestion Why should I be moral? with a surprisingly simplified response saying that such a uestion is like asking a fish why it should learn to swim Churchland says we should be moral to be a successful moral agent thus skipping over the eons long debate about what it means to be moral He adds in his last paragraph that we should make mutual love a deeper and widespread human achievement That's easier said than done and the pertinent uestion is the underlying affective mechanics that promote that objective or promote deviation from it That gets into the business of whether our souls are good or bad and how this pushes the use of the brain and reason one way and not another Churchland euates reason with the soul but these should not be conflated Understanding the biological mechanisms of brain function is a good thing That part is the engine of reason But reason doesn't function without an energy source Caring one way or another is the business of the soul and this book seems to be about reason than the soul


  5. says:

    I've never been disappointed by anything either of the Churchland's have to say I think my only real criticisms of this book have to do with how long ago it was written 15 years in the field of neuroscience is like a century in many other sciences Much of Paul Churchland's speculations regarding how different brain subsystems function have been researched in the years since this book was written fMRI and MEG are only briefly mentioned in the last few pagesThe one criticism I have that doesn't revolve around the text being dated has to do with his faith in neural networksvector coding to explain all there is to know about consciousness I am certainly no dualist but there remains a currently intractable problem regarding how these networks can substantiate meaning and subjective experience These uestions will remain out of the reach of neuroscience if we look to answer them solely through the neuronal firing underlying human consciousness and behavior In general though I applaud Churchland's method of explaining everything from sensory representations to moral behavior through vector coding and pattern recognition I'll probably check out his newer work Neurophilosophy at Work to see what an updated Paul Churchland has to say about this same subject


  6. says:

    While the ending chapter was a bit idealistic with the application of recurrent networks the book portrays the new framework of mind incredibly well There is lots of very good evidence supporting Churchland's view that he brings up in the book and still acknowledge the shortcomings we have yet to address I liked it a lot


  7. says:

    Without being able to posit a closed system for the mind Churchland's work isn't completely definitive but it's the best neurology oriented book on consciousness out there


  8. says:

    I am learning about the brain


  9. says:

    Damn good