Home » Intelligent Design » Deep Blue Never Is (Blue, That Is)

Deep Blue Never Is (Blue, That Is)

In the comment thread to my last post there was a lot of discussion about computers and their relation to intelligence.  This is my understanding about computers.  They are just very powerful calculators, but they do not “think” in any meaningful sense.  By this I mean that computer hardware is nothing but an electro-mechanical device for operating computer software.  Computer software in turn is nothing but a series of “if then” propositions.  These “if then” propositions may be massively complex, but software never rises above an utterly determined “if then” level.    This is a basic Turing Machine analysis. 

This does not necessarily mean that the output of computer software is predictable.  For example, the “then” in response to a particular”if” might be “access a random number generator and insert the number obtained in place of the variable in formula Y.”  “Unpredictable” is not a synonym for “contingent.”  Even if an element of randomness is introduced into the system, however, the way in which the computer will employ that random element is determined. 

Now the $64,000 question is this:  Is the human brain merely an organic computer that in principle operates the same way as my PC?”  In other words, does the Turing Machine also describe the human brain ?  If the brain is just an organic computer, even though human behavior may at some level be unpredictable, it is nevertheless determined, and free will does not exist.  If, on the other hand, it is not, if there is a “mind” that is separate from though connected to, the brain, then free will does exist. 

This issue has been debated endlessly, and I refer everyone to The Spiritual Brain for a much more in depth analysis of this subject.   For my purposes today, I propose to approach the subject via a very simple thought experiment. 

First a definition.  “Qualia” are the subjective responses a person has to objective experience.  Qualia are not the experiences themselves but the way we respond to the experiences.  The color “red” is the classical example.  When light of wavelength X comes into my eye, my brain tells me I am seeing the color red.  The quale (singular of “qualia”) is my subjective experience of the “redness” of red.  Maybe the “redness” of red for me is a kind of warmth.  Other qualia might be the tanginess of a sour taste, the sadness of depression, etc.

Now the experiment:  Consider a computer equiped with a light gathering device and a spectrograph.   When light of wavelength X enters the light gathering device, the spectrograph gives a reading that the light is red.  When this happens the computer is programmed to activate a printer that prints a piece of paper with the following statement on it “I am seeing red.”

I place the computer on my back porch just before sunset, and in a little while the printer is activated and prints a piece of paper that says “I am seeing red.”

 Now I go outside and watch the same sunset.  The reds in the sunset I associate with warmth, by which I mean my subjective reaction to the redness of the reds in the sunset is “warmth.”

1.  Did the computer “see” red?  Obviously yes.

2.  Did I “see” red.  Obviously yes.

3.  Did I have a subjective experiences of the redness of red, i.e., did I experience a qualia?  Obviously yes.

4.  Did the computer have a subjective experience of the redness of red, i.e., did it experience a qualia?  Obviously no.

Conclusion:  The computer registered “red” when red light was present.  My brain registered “red” when red light was present.  Therefore, the computer and my brain are alike in this respect.  However, and here’s the important thing, the computer’s experience of the sunset can be reduced to the functions of its light gathering device and hardware/software.  But my experience of the sunset cannot be reduced to the functions of my eye and brain.  Therefore, I conclude I have a mind which cannot be reduced to the electro-chemical reactions that occur in my brain.

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

188 Responses to Deep Blue Never Is (Blue, That Is)

  1. Jerry: Thanks for @153

  2. StephenB,

    You’re welcome but I have to thank you for the great post.

  3. KF, in 180, mentions about brain, mind, interface and information “We can discuss these separately, but we can’t design, develop or put them together separately.”
    Depends upon how one parses the problem. Information can be implemented without interfacing to mind – at least the properties of information as it resides in the brain. Interface can be examined without information flowing across it – at least the physical brain-side of the interface, such as using cadavers with no information flow. Just like any well-structured computer problem, the modules can be examined in isolation from the others.

  4. Q:

    Re: Information can be implemented without interfacing to mind . . . Interface can be examined without information flowing across it

    We are speaking to a particular context and thus also the DS architecture.

    It is moreover the case that interfaces are designed in the context of the requisite information flows, thus the informaitonin question.

    Here we are interested in the predictive paths set up imaginatively, volitionally and creatively for the effecting servo-system. These serve as templates for action. What is significant is that relative tot he control part, the templates are in effect givens ands the controller can syntactically track and compare actual with projected then generate control vectors to correct deviations.

    The setting up of the track and of the programs to control the controller are SEMANTIC. That is, for instance a bit based processor can simply sample outputs at given times then compare to expected, generate error signals that drive actuators and monitor onward performance. But what the signals mean is not a necessary part of that — it is in the semantics. What the controller is doing is register-based arithmetic, logic and shift operations on bit strings; it does not itself address what the strings mean.

    That is the job of the programmer.

    As people we plan then act and we monitor deviations and respond to them to get back on track. We do so intelligently – based on meanings and what makes sense, as a rule.

    GEM of TKI

  5. Pardon my ignorance, KF, but I miss your point in 184. I’m discussing how the computer model you present can be used used as a tool to examine the brain/mind duality. I.e., in the context of this thread, to observe the boundary between “seeing red” and “subjectively experiencing red.”

    I’m arguing that the model you present must be correlated to the physical brain, they physical interface, and the physical information – through observation of the brain, its interaction with information, and its interaction with the interface to the mind. The model you present must not simply treated as a direct representation of the brain. For this examination, brain, information, interface, and mind, and be treated and observed as separable constructs.

  6. Q:

    In my context, I am making no initial distinction on whether what we experience as the mind is material or immaterial at the first instance. I am simply pointing to the DS architecture, that allows us to differentiate controller from intelligent director and to assign the locus of creative tasks – getting beyond the Crick-style confusion.

    Then, let us bring to bear the relevant issues on what an intelligent director would be like and how it is set up, by going back to a point in my always linked, section A and a remark by good old materialism-leaning prof Wiki on Instincts [and along the way, DV, we will make reference again to Ac 27 on governance by competing agents in a situation that exhibits tracking in the short term and navigation in the long term relative to an intended path]:

    [GEM of TKI:] let us identify what intelligence is. This is fairly easy: for, we are familiar with it from the characteristic behaviour exhibited by certain known intelligent agents — ourselves. Specifically, as we know from experience and reflection, such agents take actions and devise and implement strategies that creatively address and solve problems they encounter; a functional pattern that does not depend at all on the identity of the particular agents. In short, intelligence is as intelligence does. So, if we see evident active, intentional, creative, innovative and adaptive [as opposed to merely fixed instinctual] problem-solving behaviour similar to that of known intelligent agents, we are justified in attaching the label: intelligence. [Note how this definition by functional description is not artificially confined to HUMAN intelligent agents: it would apply to computers, robots, the alleged alien residents of Area 51, Vulcans, Klingons or Kzinti, or demons or gods, or God.] But also, in so solving their problems, intelligent agents may leave behind empirically evident signs of their activity; and — as say archaeologists and detectives know — functionally specific, complex information [FSCI] that would otherwise be improbable, is one of these signs.

    [“prof” Wiki, 1:] Instinct is the inherent disposition of a living organism toward a particular behavior. Instincts are unlearned, inherited fixed action patterns of responses or reactions to certain kinds of stimuli. Innate emotions, which can be expressed in more flexible ways and learned patterns of responses, not instincts, form a basis for majority of responses to external stimuli in evolutionary higher species, while in case of highest evolved species both of them are overridden by actions based on cognitive processes with more or less intelligence and creativity or even trans-intellectual intuition.Examples of instinctual fixed action patterns can be observed in the behavior of animals, which perform various activities (sometimes complex) that are not based upon prior experience and do not depend on emotion or learning, such as reproduction, and feeding among insects. Other examples include animal fighting, animal courtship behavior, internal escape functions, and building of nests.

    Instinctual actions – in contrast to actions based on learning which is served by memory and which provides individually stored successful reactions built upon experience – have no learning curve, they are hard-wired and ready to use without learning, but do depend on maturational processes to appear.

    [PW, 2:] Intelligence is an umbrella term used to describe a property of the mind that encompasses many related abilities, such as the capacities to reason, to plan, to solve problems, to think abstractly, to comprehend ideas, to use language, and to learn. There are several ways to define intelligence. In some cases, intelligence may include traits such as creativity, personality, character, knowledge, or wisdom.

    [PW, 3:] Creativity (or “creativeness”) is a mental process involving the generation of new ideas or concepts, or new associations between existing ideas or concepts.

    From a scientific point of view, the products of creative thought (sometimes referred to as divergent thought) are usually considered to have both originality and appropriateness. An alternative, more everyday conception of creativity is that it is simply the act of making something new.

    [PW, 4:] Intuition is apparent ability to acquire knowledge without a clear inference or reasoning process.

    It is “the immediate apprehension of an object by the mind without the intervention of any reasoning process” [Oxford English Dictionary].

    Intuition, by definition, has no objective validity. However it is extremely widespread as an apparent phenomenon. For this reason, it has been the subject of study in Psychology, as well as a topic of interest in the supernatural. . . . In psychology, intuition can encompass the ability to know valid solutions to problems and decision making. For example, the recognition primed decision (RPD) model was described by Gary Klein in order to explain how people can make relatively fast decisions without having to compare options. Klein found that under time pressure, high stakes, and changing parameters, experts used their base of experience to identify similar situations and intuitively choose feasible solutions. Thus, the RPD model is a blend of intuition and analysis. The intuition is the pattern-matching process that quickly suggests feasible courses of action. The analysis is the mental simulation, a conscious and deliberate review of the courses of action

    These — together with the DS architecture of a complex servo-system with a controller based on input-output comparison to projected track, and with the projected track being creatively supplied by what I have called an intelligent director – will form a context for the further remarks. [Then, we can deal with subjectivity, consciousness and qualia etc as markers that point to the nature of the relevant director we possess.]

    1] Directors and neural network characteristics and programming.

    In a sense this reworks what was dealt with under a similarish post on a parallel thread, but with adjustments to this thread.

    For, we know what agency is, DIRECTLY IN THE FIRST PERSON, so we experience that intuition, creativity and intelligence are features of agency that routinely act effectively into the world. This is what has to be reasonably explained.

    Thence, we can look at the DS framework and the relevance of an intelligent director – or of a collective of such directors [per Ac 27] — supervising and guiding the i/o processor controlling the servosystems: robot of the future, body in the present, or ship in the past of October 59 AD makes little difference.

    2] Neural networks as a model . ..

    Wiki on neural networks:

    in unsupervised learning [in a neural network] we are given some data x, and a cost function to be minimized which can be any function of x and the network’s output, f. The cost function is determined by the task formulation. [ note this — someone sets the task, sets the goal and sets up the system, i.e the ANN does not ultimately question its final-level purpose.] Most applications fall within the domain of estimation problems such as statistical modeling, compression, filtering, blind source separation and clustering . . . .

    In reinforcement learning, data x is usually not given, but generated by an agent’s interactions with the environment. At each point in time t, the agent performs an action yt and the environment generates an observation xt and an instantaneous cost ct, according to some (usually unknown) dynamics. The aim is to discover a policy for selecting actions that minimises some measure of a long-term cost, i.e. the expected cumulative cost. [Note the preset purpose.] The environment’s dynamics and the long-term cost for each policy are usually unknown, but can be estimated. ANNs are frequently used in reinforcement learning as part of the overall algorithm. Tasks that fall within the paradigm of reinforcement learning are control problems, games and other sequential decision making tasks.

    Notice how the learning control system has to be set up to have a creative, imaginative, intelligent and even intuitive supervisory view of the world and its dynamics and conditions so that it can explore and address then model potential costs and benefits of policies then go for the goals it has, and of course monitor and adjust as it tracks across time, leaning from experience.

    That brings up the governance issue as competing policies vie for adoption. Thence, Acts 27 and issues of democratic governance and wisdom.
    But this is a bit afield. We want to go now for what agents are like in our own case.

    3] Mind-brain issues — simplified

    As BarryA observed in the OP:

    computer hardware is nothing but an electro-mechanical device for operating computer software. Computer software in turn is nothing but a series of “if then” propositions. These “if then” propositions may be massively complex, but software never rises above an utterly determined “if then” level . . . . the $64,000 question is this: Is the human brain merely an organic computer that in principle operates the same way as my PC?” In other words, does the Turing Machine also describe the human brain ? If the brain is just an organic computer, even though human behavior may at some level be unpredictable, it is nevertheless determined, and free will does not exist. If, on the other hand, it is not, if there is a “mind” that is separate from though connected to, the brain, then free will does exist . . . .

    “Qualia” are the subjective responses a person has to objective experience. Qualia are not the experiences themselves but the way we respond to the experiences . . . . Consider a computer equiped with a light gathering device and a spectrograph. When light of wavelength X enters the light gathering device, the spectrograph gives a reading that the light is red. When this happens the computer is programmed to activate a printer that prints a piece of paper with the following statement on it “I am seeing red.”

    I place the computer on my back porch just before sunset, and in a little while the printer is activated and prints a piece of paper that says “I am seeing red.”

    Now I go outside and watch the same sunset. The reds in the sunset I associate with warmth, by which I mean my subjective reaction to the redness of the reds in the sunset is “warmth.” . . . .

    Conclusion: The computer registered “red” when red light was present. My brain registered “red” when red light was present. Therefore, the computer and my brain are alike in this respect. However, and here’s the important thing, the computer’s experience of the sunset can be reduced to the functions of its light gathering device and hardware/software. But my experience of the sunset cannot be reduced to the functions of my eye and brain. Therefore, I conclude I have a mind which cannot be reduced to the electro-chemical reactions that occur in my brain.

    Actually, I am astonished that we have to go down to so many details to see the obvious. There are many current views, that (like Crick), would reduce mind to brain. But from our experience of mind – which is necessarily relied upon to think even materialistic thoughts – we do experience free will and intelligent creativity, intuition etc.

    Even in the case of learning artificial neural networks, they have to be set up in ways that fairly reek of organised complexity, pointing onward to agency and intelligence. And, free thinking and acting are conditions of such intelligence. Further to this, we experience ourselves as such intelligent agents.

    Thus, plainly, any view that contradicts the facts of intelligent agency as we experience it, are false-to-fact, and falsified. [IMHBCO, it is the institutional power of lingering evo mat that makes this hard to do, not the logic.]

    And, those facts plainly contradict the notion that mind is an emergence from the properties of matter as we understand them through scientific study. So, on the evidence in hand, mind is more than mater but is capable of interacting with it in interesting ways. Most notably, it is capable to provide the creative, imaginative, intuitive etc that can then guide the servostysems involved.

    4] And if mind has been created . . .

    Then prima facie, mind can be created.

    So, R Daneel is in principle possible. The issue is: how!

    So, let’s roll up our sleeves and sharpen our pencils – the adventure of design science has only just begun . . .

    GEM of TKI

  7. I think I see the difference in our approaches, and that it may result in different outcomes. Your approach, if I read it correctly, is to start with the Intelligent Director. Specifically, you start with “(L)et us bring to bear the relevant issues on what an intelligent director would be like and how it is set up”

    That is, the Intelligent Designer is one of your premises.

    I’m taking a different point. Start with the measurables – brain, information flow, brain-side of the mind/brain interface (or the correlaries in the DS model) – and by a process of elimination, conclude with what are the properties of mind (or of the Intelligent Designer). My approach makes no initial assumptions about the properties of each element of the process, expect for the assumption that a particular model is being followed, as your logical argument suggests. However, by observing the processes and boundaries of each element, we would conclude with the understanding of “that which remains only indirectly observed is left as mind (or as the Intelligent Director)”.

    If the model were correct, would you expect both approaches to result in the same understanding of the Intelligent Designer, or of mind?

  8. Q:

    I just now spent a significant amount of time on a response to Prof Olofsson, harking back to the Padian thread. So pardon my being a bit summary, especiaslly as I see us looping back over old ground.

    1] 187: the Intelligent Designer is one of your premises.

    This loops back to the objection of the Kantians, and is in serious error, as I long since pointed out and linked.

    In the current context, I am looking at a MODEL, by DS, in which what I have called an Intelligent Director is a part, the part that passes creative projected paths for a servo to the controller, whose job is to then try to keep the system on track from moment to moment thus executing the path desired. It is the possession of such an intelligent director capable of making such decisions and creative projections that makes the model in Fig 2 in the predictive form, self-directing and capable of praxis.

    Notice the difference in context and term,s, please. On inference to design, what the explanatory filter approach does is to refuse to rule out by begging the question the possibility of a designer at OOL OOBPLBD and OO FT LFC. Then, the strongly empirically and theoretically supported point that FSCI etc are reliable signs of agency is seen as pointing to agency on a basis of inference to best inductively anchored explanation.

    The price tag for rejecting this is selective hyperskepticism, as I have pointed out.

    2] Start with the measurables – brain, information flow, brain-side of the mind/brain interface (or the correlaries in the DS model) – and by a process of elimination, conclude with what are the properties of mind

    First, one measures in a context that already implicitly addresses explanatory alternatives, i.e the models are there all the time.

    Second, on our own case we start from our life-experience as agents. We know from the inside what it is to be intelligent, creative etc, and that is the context in which mind as a term has been developed. We are therefore reasoning by family resemblance to known cases in point. Whatever mind is, we have an example in point that is empirical.

    We now set out to see what “stuff” mind is made of – what it is is different from that it is; the latter being the premise of all our intellectual activity. In that pursuit, we see that certain artificial entities are a possible model: supervised servosystems in which an intelligent director creatively sets the path and the expected observations along it to guide the controller to keep on track.

    We can compare two cases, one that would be hard to do in realistic cases [we have already done simpler case by making Model-referenced adaptive controllers and their descendants] but is technologically feasible, and one [Ac 27] where we see a classical account of a ship voyage under direction of a steersman in the face of decisions by the ship’s company and the environment.

    The result is to show that information is a key intermediary between intelligent direction and control.

    Further by comparison with our experience we know that creative synthesis of such paths is based on understanding of configurational possibilities and dynamics, thus on going right to islands of functionality instead of being lost in vast config spaces and trying to find function through random walks that soon run out of probabilistic resources. On this, the neural network model is useful, but notice the types that speak to such self-directed learning tare also involved in serious FSCI to set them up.

    So we see a crucial difference between mind and chance + necessity only in design.

    3] My approach makes no initial assumptions about the properties of each element of the process, expect for the assumption that a particular model is being followed, as your logical argument suggests. However, by observing the processes and boundaries of each element, we would conclude with the understanding of “that which remains only indirectly observed is left as mind (or as the Intelligent Director)”.

    A look at the just above will show that it reiterates that there is no assumption before the fact on the ontological nature of mind, only that we recognise that mind is – we ourselves experience it.

    More to the point we also infer that since mind in our case is contingent, creation of mind is possible so the trick is to identify how to do it. Thence R Daneel here we come!

    By observing he characteristics of the DS model andf the relevant neural networks and comparing with the observed and experienced behaviour of human minds, we then can make comparisons on boundaries and ask questions on the origins and ontology of the relevant components.

    GEM of TKI

Leave a Reply