It has struck me more than once that a lot of the confusion that accompanies discussions about stuff like consciousness and free will, and intelligent design, and teleology, even the explanatory filter, and the words “chance” and “random” arises from lack of clarity over the difference between decision-making and intention. I think it’s useful to separate them, especially given the tendency for people, especially those making pro-ID arguments, but also those making ghost-in-the-machine consciousness or free will arguments, to regard “random” as meaning “unintentional”. Informed decisions are not random. Not all informed decisions involve intention.
This was my first computer:
It was called Dr Nim. It was a computer game, but completely mechanical – no batteries required. You had to beat Dr Nim, by not being the one left with the last marble, and you took turns with Dr Nim (the plastic board). It was possible to beat Dr Nim, but usually Dr Nim won.
Dr Nim was a decision-making machine. It would decide how many marbles to release depending on how many you had released. Frustratingly, it seemed no matter how clever you got, Dr Nim nearly always left you with the last marble. Here is a youtube demonstration:
Clearly, Dr Nim is not acting “randomly”. It wins far more often than would a random system that selected 1, 2, or 3 marbles regardless of how many there are left. In fact, it seems to “know” how many marbles there are left, and chooses the best number to drop, accordingly. In other words, Dr Nim makes informed decisions. But clearly Dr Nim is not an “intentional agent”. It’s just a few plastic gates mounted on a channeled board.
And yet Dr Nim behaves like an intelligent agent. It was certainly smarter than me at playing Nim!
I suggest that the products of evolution look like the products of intelligence( as in informed, non-random, decision-making) because they are the products of intelligence (as in informed, non-random, decision-making). The mistake I think ID proponents make it to assume that such a system must be intentional.
What’s the difference? I suggest that an intentional decision-maker is one that is able to model a distal goal, and select actions from a range of possible actions, on the basis of which is most likely to bring about that goal. And I suggest that humans, for example, do this by simulating the outcomes of those actions, and feeding the results of those simulations back into the decision-making process. This allows us to cut corners in a way that evolutionary processes can not, and evidently, do not. It also, I suggest, gives us enormous freedom of action – as in “degrees of freedom” – not to do “random” things (which would be the opposite of “intentional”) but things that we will – intend. Although sometimes it makes us not quite as clever.
Then evolution has a goal?
What goal would that be?
The mistake I think you make it to assume that ID proponents assume such a system must be intentional. They do not, because obviously Dr. Nim and more advanced computers exist, and I doubt you believe that ID proponents assume such machines are intentional.
What an ID proponent might argue, however, is that such machines (some, all, or those beyond a certain metric threshold) are probably the result of intentional planning towards a distal goal.
I’ve had a discussion with several ID proponents about this, and also read Dembski very carefully on this, and certainly some do. Dembski oddly, defines intelligence in such a way that excludes intention, but, thereby, includes evolution. He, however, considers, wrongly, that “intention” is a necessary inference, but outside the domain of science.
As long as we exclude “intention” from our definition of Intelligent Design, I am an ID proponent. But most people think of “intelligence” as being necessarily “intentional”, and the word “Design” as capturing the intentional part (“did you do this accidentally, or was it by design?”). And many IDists use “random” and “unintentional” interchangeably (Cornelius Hunter, for one).
That’s why I attempt to separate the two concepts in this post, and use the word “decision-making”, instead.
I do think that evolution is a very sophisticated decision-maker. I don’t think it is an intentional decision-maker.
I want a Dr. Nim! What a great device.
Lizzie – I agree with most of what you say but I think you can take it a step further. The definition of “random” is famously difficult but one reasonable definition is “unpredictable” and that given that definition it is matter of degree – things can be more or less predictable – then you have two dimensions about any decision: predictable/random and intentional/unintentional. I propose that any of the resulting four combinations are possible:
random/intentional e.g human conscious decision making
random/unintentional e.g. lottery
predictable/intentional e.g. less intelligent animal decision making
predictable/unintentional e.g. Dr.Nim
This raises a couple of interesting cases:
(1)
You write
“It also, I suggest, gives us enormous freedom of action – as in “degrees of freedom” – not to do “random” things (which would be the opposite of “intentional”) but things that we will – intend.”
I don’t think random is the opposite of intentional. Simulating outcomes may make my decision more predictable but there could still be a large random component both in the range of outcomes that I happen to simulate and in which one I eventually go for after the simulation.
(2)
A chess playing computer has a model of future outcomes and choose between them. I would have no problem as describing this as intentional but I wonder if you are happy with it?
Yes, I think a chess-playing computer could be described as “intentional” by my definition – it models the possible outcomes of various future moves and selects the one most likely to produce the goal outcome.
I agree that “random” is not the opposite of “intentional” in most usages, but I think some inadvertent (possibly advertent) equivocation creeps in during a lot of ID discussions. Cornelius Hunter, for example, repeatedly describes the “Darwinian” explanation as “random”, and no matter how much people point out that natural selection is not “random” in many senses of the word (though it is stochastic), he insists that it is, for example in this post:
And later, in the comments, says:
So he does seem to be equating “random” with “undirected”, by which he seems to mean “unintentional”.
So there is definite confusion here, and I’d say Hunter is not alone.
I think another thing that Dr Nim demonstrates very nicely is that “knowledge” can be embodied in a configuration of switches. I think the same is true of brains. We have had discussion here about “brain states” “representing” knowledge. I don’t think that Dr Nim’s switch state “represents” knowledge (except to an external observer trying to figure out what Dr Nim is doing!). Dr Nim’s switch state is the knowledge, specifically, of what to do next given what has gone before. In other words, knowledge, as I keep saying, as embodied in brain states, preparation for action. It doesn’t “represent” it.
Now I do think that we also “represent” knowledge to ourselves. But those representations are not our brain states.
(Also, Dr Nim neatly disposes of Upright Biped’s semiotic nonsense I think but that’s another matter :))
Liz said:
Intention is the essential component of intelligent design, so no, I don’t think anyone will agree to that. Since I don’t think anyone believed we achieved “artificial intelligence” with the invention of Dr. Nim, I think it is you that is using a historically and conventionally inaccurate term when you insist that computation = intelligence. Also, I think you are doing this (consciously or unconsciously) to subvert the ID argument by definitional fiat by redefining “intelligence” as “computation”.
Dr Nim computes; it is not intelligent. Computers compute. They are not intelligent.
Instead, we call what Dr. Nim does “computaton”, and what humans do “intentional computation” .. or intelligence.
I think that you are once again attempting to subvert a definition, much like your use of the term “intelligence”, into meaning something which will be convenient to your metaphysics.
Okay, so you are subverting the definition.
Perhaps you could explain how theses statements are reconciled, given that a “model” essentially means “a representation of something”. If representations are not states of the brain (or states of a computer), how does a chess-playing computer model possible outcomes and future moves?
You say that a computer is able to model a distal goal; how does a computer go about choosing a distal goal, unless there is a representation of what that distal goal “means” somewhere in the information on its circuitry, hard drive or RAM (“brain states”)? How does a computer choose the criteria by which it chooses goals, or upon having selected a goal, the criteria by which paths to the goal are judged? If “intention” is “modeling a distal goal”, then what commodity chooses goals, and chooses the means by which to judge pathways to the goal?
Modeling paths to a goal is largely just computation after the intention has chosen a goal, and after intention has defined how paths to the goal will be judged/generated (according to what criteria). Intention can also change or modify the goal, or how paths to the goal are being judged/generated, if new things occur to the intentional agent during the process, either as part of feedback or from outside of the goal system in question.
Can a chess-playing computer choose what a “perfect” apple fritter would look like, and then begin modeling a path to the goal? Can a chess-playing computer change the definition of “winning” in chess to “whatever I say winning means” so that it can win simply by definitional fiat?
Mark Frank said:
If “random” = “unintentional” as Liz suggests (and I agree), the the category “random/intentional includes both X and not-X. I suggest that there is one “dimension” to decision-making, with three possibilities: predictable, random (or, as Liz might prefer, not yet predictable), and intentional.
Corresponding to necessity, chance, and artifice.
I would also like to point out that in the examples of Dr. Nim and a chess-playing computer, the “distal goals” are determined by an outside agency that is capable of choosing whatever goal that outside agency wishes, and then develops physical states and systems that fulfill that goal.
Would one expect a working, functioning Dr Nim to come into existence without the help of an intentional agency? Yes or no?
Yes indeed. And I think it is very important to make exactly that distinction – between the internal goals of a decision-making system, and the goals of whoever, or whatever, produced the system.
Probably not, but it’s so simple, a billion monkeys with bits of plastic just might manage it. However, my favorite example is system that produced and maintains Chesil beach, which is not obviously the result of an intelligent agency, and yet is a magnificent sorting machine.
Be that as it may, right now, I am trying to make a distinction between a system that makes informed decisions, and a system that makes intentional, informed decision. I’d say Dr Nim is in the former camp, and Mark’s chess machine is in the latter (by my definition). Neither, are conscious, I’d say, though.
That’s yet another thing 🙂
But both are designed by an agent capable of conscious, intentional, informed choice.
Lizzie you have a problem with what is free will and what is a decision. Dr. Kim don`t do any decision.
Because Dr. Kim has not will. Dr. Kim do not “know” how many marbles you have taken. It is just a device intended to solve mechanically a mathamatical problem that is he Dr. Kim game.
The game will be won always by the second player if answer taking an odd number when the first has taken two marbles and a none number of marbles when the first player has taken one or three.
What does the machine? When the first player start to chose marbles the first marble the keys in the way the marble follows are turned in a position that if the players give to Dr. Kim the turn the next marble will go without changing the turn of play, so the answer of Dr. Kim to the choose of one marble by the first player will be the winning two marbles. The second marble taken by the first player, turns the keys in order that if the next turn will be for Dr Kim one marble will make Dr. Kim to return the game to the first player, so the answer of Dr. Kim to the choose of two marbles by the first player will be the winning one marble. If the first player choose three marbles again the keys are change in order the answer will be two marbles.
As you see, there is no “knowledge” of Dr. Kim, and definitely there is no will of Dr Kim, Dr. Kim do not acts, the law of physics make things happens. You can foul Dr. Kim using an none number of marbles and you will win. Dr. Marble will not know and will not be able to change the strategy, you can change the positin of the keys manually an Dr. Kim will not react.
Let the question that IDist makes Could a process of RM+NS build a Dr. Kim machine? Because that queston is easy to answer for you materialist. If man is a product of RM+NS, man builded Dr. Kim, then RM+NS builded Dr. Kim.
The problem for materilist is, as Dr. Coyne realize, that RM+NS can build only machines like Dr. Kim.
When the level of glucagon in the lion`s tissues go dows, the “keys” of his organisms makes the lion wlaks, his internal keys make him follow the smell of a pray, his internal keys make follow the most easy pray, run, kill an eat. Then the keys are restored and the lion goes to sleep. We say that he got hungry and he “wanted” to eat, but the lion as us are only Dr Kim machines with marbles changing the position of keys.
So not the “free” will is an illusion, the “will” is an ilusion and the question that Dr. Coyne has to answer is why a RM+NS process will build a Dr. Kim machine that autoilludes to have a will, a free will, propose a mathematical game, solve the mathematical game, build a machine that plays the game and knows how to fool the machine.
You can define words however you want, Liz. IMO, however, what you are doing is deceitful, in the same way that the phrase “compatibalist” free will is deceitful. IOW, using “intelligence” and “intentional” the way you do does nothing more than begs the question, moving the essential nature of the term outside of the new definition and back a step.
IOW, what you define as “intelligence” and “intention” can as easily, and more conventionally, be served by using the term “computation”, and IMO the only reason (conscious or not) to co-opt terms like intelligence and intention (and even “decision”, IMO) as, essentially, “forms or parts of computation” is to obfuscate and obstruct fair debate and conceptual clarity about the issue.
IMO, it all smacks of avoiding the real debate by undermining the lexicon those who disagree with you use to explain and argue. If you redefine the essential words they use to make their case as meaning something other than what they are using them to denote, then you’re simply rigging the debate in your favor by “fixing” definitions conveniently.
Why not just use the word “compute”, instead of “intelligence”, or “intention”? Why must you reach across the ideological aisle, take our terms, and redefine them, when you have perfectly adequate words already available that comport with your ideological position?
Here’s a question: In your opinion, does all thought, belief, intelligence and intention result from a collection of interacting, physical computations (physics), even if that computation may include variables that are indeterminate?
I’m not attempting to “subvert” anything by “definitional fiat”. What I am attempting to do is to produce an operational definition of various terms so that we can avoid equivocation. We can substitute symbols, or subscripts if you like – the important thing is that we agree, for the purposes of this discussion, what refers to what. That’s why I used “decision-making”. We can call “intelligent”, a “decision-maker”, or an “intentional decision-maker” or a “conscious, intentional decision-maker”. I don’t mind. But it’s important to know what we mean when we use the word, and what the person we are talking to means. I will try to avoid the word where possible, and use the above instead.
Regarding ID, however, Dembski at one stage provided this operational definition: “by intelligence I mean the power and facility to choose between options–this coincides with the Latin etymology of “intelligence,” namely, “to choose between””. I’d say that Dr Nim “chooses between” (aka “selects”) options, but is neither intentional, nor conscious. Whether one wants to call that “intelligence” is up to the user of the word – my point is merely that we need to specify precisely what we mean, and Dembski, who is, I believe, an ID proponent :), appears to regard “intention” as extrinsic to his definition. In fact he also states, later in that same paper, at the end of a list of “key problems” that should be addressed “once it is settled that certain biological systems are designed”:
In other words he regards intention (wrongly, in my view) as not only not a question of science (I think it is) but as something that we consider once we have established intelligent design. In other words, according to Dembski, intention implied by decision-making, it is not part of the definition. I think Dembski has made an error here.
But whether he has or not, I think it’s important to be clear that a non-intentional system can nonetheless make informed decisions. Dr Nim is capable of selecting one, two, or three marbles, and which it selects depends on its internal state, which in turn depends what the other player does.
I’m not making any grand claims for Dr Nim; it’s just a very nicely designed (by human beings) piece of plastic. My only claim is that it makes informed (not “random”) decisions. Computers do the same – you put in a question and it spits out an answer. The answer is not random – the answer depends on the input, and we frequently get answers that we cannot know in advance – those answers depend on a cascade of binary decisions within the computer.
I’m more than happy to regard a computer as not “intelligent” – to define, in other words, “intelligent” as requiring conscious intention. But if we do define intelligence thus, we need to be clear that there are sub-intelligent decision-making systems that can produce highly “non-random” output – that can, in other words make informed decisions.
This is important.
William J. Murray,
Calling Lizzie deceitful is rich, William, when certain other people avoid defining their terms at all costs. What can you offer as a definition for “intelligence”?
Something being avoided is the concept of learning. Nim is completely deterministic, but it is possible for a deterministic substrate to support a system that modifies itself as a result of feedback and which behaves in unpredictable ways.
No, it is not “deceitful”, William, which would, as I’m sure you agree, imply an intent to mislead. Such is not my intent. My intent is quite the reverse – to make clear. I think yet again (and this is not meant to pull rank in any way, it is just an observation) that this perception arises from the culture gap between science and non-science. In science we have to define our terms extremely carefully, hence the term “operational definition”. When we do this, the idea is not to declare, by “definitional fiat” as you put it, that for all time “X” means “Y”. It is to say, for the purposes of this discussion, let X denote Y.
And what I am doing is the very opposite of begging the question, or equivocating – equivocation is precisely what I am trying to prevent – from using the word “intelligent” or “random” in one sense in one context, then carrying over the same definition into a different context. This is what I think Cornelius Hunter is doing (inadvertently I’m sure) and I’m trying to lay out the parts of the concepts, and label them clearly differently. What labels we use is irrelevant. But I’m trying to use labels that carry as little former baggage as possible.
Hence my use of the terms “decision-maker”; “intentional decision-maker” and “conscious intentional decision-maker”. What matters is how I have defined them for the purpose of this discussion
I’m happy to use “computation” as long as you give me a very clear operational definition of what it means. In lieu of that, I will stick with what seems to me clear, which is “decision-maker”.
No. It’s actually the exact opposite. I think you are reading into my words something that isn’t there. You suspect me, I think of saying: look Dr Nim is intelligent, therefore evolution is intelligent, therefore evolution = intelligent design.
No, I’m not.
I’m saying: decision-makers need not be intentional or conscious, to produce informed, rather than “random” decisions. Cornelius Hunter implies that because evolution is alleged to be non-intentional (and I would agree), it can therefore be described as “random chance”. I’m saying, no, that doesn’t follow. There is a middle, for which perhaps we lack a label (but which, oddly, meets Dembski’s definition of “intelligence”, although not mine) that is excluded if we define all non-intentional systems as “random chance”. Dr Nim is one. I suggest evolution is another.
I took Dembski’s word “intelligence” as he defined it But I am more than happy to use a different one (I don’t think his is very good – I think it doesn’t capture what we normally mean by the word, and doesn’t even capture what he appears to mean by the word). All I ask is that people (including Cornelius) define their terms, and use them consistently in a given conversation.
If you give me an operational “computation” I’d be happy to answer 🙂 Not a “definition by fiat”, but simply a definition that we can use for the purposes of addressing your question.
Absolutely. I frequently model learning on a computer, and although I use a random number seed, if I set the same seed each time, I will get the same results. The program learns, whether or not I reset the random number seed, although the time it takes may vary, and the solutions it comes up with may also vary.
How do you know whether or not you have an intent to mislead, Liz? Are you saying that in order to intend something, you must be aware of your intention?
Aren’t we using your definition of “intention” here, where you claim that “modeling pathways to reach a goal” = intention, and that a chess computer is an intentional agency? Where did you say anything about “awareness” in that definition?
Are you claiming that a chess computer is aware of what it is doing? If not, then by your definition and example, intent doesn’t require awareness of intent, and you could very well be deceitful and not know it.
Right?
No, by definition.We define lying as stating something knowingly untrue.
Intelligence: capacity to reason, to plan, to solve problems, to think abstractly, to comprehend ideas, to use language, and to learn.
Amen!
I disagree that any non-intentional agency makes “informed decisions”; IMO that term is deceitful, whether deliberately so or not.
What non-intentional agencies do is compute (in terms of physics) what necessarily follows from what precedes based on it’s structure/programming. It no more “makes a decision” than a rock rolling down a hill “makes a decision”, nor does it make an “informed” decision simply because of the information contained in its shape, physical structure, composition and mass as it computes a path down a mountainside.
Water “computes” (materials acting according to physics) its path down the hill; it doesn’t make informed decisions. Setting up a system of channels and water-weight switches doesn’t give the water or the system any additional decision-making power. The water still computes according to physics. The meaningful “decisions” have already all been made by that which created the channel system which takes advantage of the natural, physics-based computing power of water (and, of course, gravity, etc.)
The “Decisions” being made by Dr Nim are being made, actually, by the person who invented it, the person that decided what Dr Nim would do in each possible scenario.
Calling that “making an informed decision” on the part of Dr. Nim is nothing but a demonstration of the absurdity of your definitions.
Lizzie,
“The answer is not random – the answer depends on the input, and we frequently get answers that we cannot know in advance – those answers depend on a cascade of binary decisions within the computer.”
It is determined by how the cascade of binary decision was build. So it will be always the same given the same input will give the same answwer.
I’m using Liz’s definitions, not whomever you are referring to as “we”. I didn’t say she was lying, I said she was being deceitful. There’s a difference. Someone can tell the truth and still be deceitful.
Well, neither is true, in this case, William.
(remainder of post moved to thread it was originally intended for!)
That’s would be true in a non-stochastic (deterministic) system. We can also build stochastic decision-makers. They often work better, because they sample more solution-space.
Not in the accepted sense of the word. This is something that disappoints me, William. You seem to have no genuine interest in communication. Surprise me by explaining how anyone can be truthful and deceitful simultaneously.
That’s a good point, William. I’d say that “deceitful” would be meaningless in a system without self- or other-awareness.
Revolting. Rather than simply apologizing for the rude mean words where he stated that Lizzie is being deceitful, he digs further to state that he’s still right if we assume, as he does. that Lizzie could perhaps be deceitful without intending to be deceitful.
Should just have the common decency to apologize and retract the “deceitful” claim. But then, common decency isn’t all that common, is it.
The general game of NIM:
The starting position is some number of piles of stones, could be any number. Each pile has some number of stones, variable.
The play is, each player in turn can take some number of stones (1 or more) from any one pile, up to the entire pile.
The winner (agreed on ahead of time) is the one who can either take the last stone, or avoid taking the last stone. For illustration purpose (as in the OP), let’s use the latter, of avoiding taking the last stone.
The calculation: express the number of stones in each pile as a binary number, and write these numbers down in a column for addition. Sum each column. If the number of ones is odd, the sum is 1; if even, the sum is 0. There are no “carries”.
If the number of ones in the sum is even, you have already lost! (against proper play, of course). Otherwise, take some number of stones from any pile such that the resulting new “sum” has an even number of ones. No matter what your opponent does (keeping the number of ones even or making it odd), simply keep removing stones to keep that number even or make it even. Eventually your opponent must take the last stone.
(And clearly, if the goal is to take the last stone, reverse all these polarities. If you’re faced with an odd number of ones in the sum, you’ve already lost, etc.)
It’s possible to lie by omission. But then you wouldn’t be “truthful” I guess. No wonder we are asked in court to tell BOTH the whole truth AND nothing but the truth.
But in any case this is irrelevant. If William thinks I am being deceitful he has not understood my point, or, indeed, the concept of an operational hypothesis. When a mathematician says: let X be the number of sheep in the field, she is not pretending that X always means “the number of sheep in a field”, and in particular, she is not trying to deceive anyone into thinking that because X means the number of sheep in field, the X-factor is a tv program about counting sheep.
My point is a very simple one: that we can have quite complex decision-making mechanisms, whether designed by human beings, like the designers of Dr Nim, or merely a result of non-living forces, as in the decision-making system that deposits small pebbles at the western end of Chesil Beach, large pebbles at the eastern end, with intermediate sizes finely graded in between, so precisely that fisherman allegedly landing at night on the beach could tell where they were in the dark by rolling the pebbles in their palms.
The result is a highly non-random pattern, i.e. one that is very unlikely to arise if the system was such that any pebble, regardless of size, had an equal probability of landing anywhere on the beach.
OK. These are all attributes the “intelligent designer” of ID theory possesses?
I don’t take offence at William, because I don’t think he intends to be offensive 🙂
Tell me if I’m wrong, William (if it is possible for you to determine this) and if so, I shall duly take offence 🙂
That seems a good all-purpose definition. It is not, however, Dembski’s. But no matter. I’m happy to use your definition, which does indeed incorporate my concept of intention (which is similar to the idea of “to plan”).
You said earlier, in response to my calling your redefinitions “deceitful”:
And now you say:
So are you saying that an entity need not be aware in order to intend?
Yes, by the definition I’ve given. But to intend to deceive, I think the deceiver needs to be aware of (be able to model) what the world looks like from the deceivee’s point of view, in other words to have some rudimentary “theory of mind” capacity. This is probably an extension of self-awareness, rather than a precursor.
These are all attributes of the only currently known ID agent – humans. That is all we currently have to base our definition of intelligence, and thus Intelligent Design, on.
Although in AI, it’s possible that other-awareness could precede self-awareness I guess. hmmm.
One word: feint. Done by preditors and prey at all levels of brain power. One of those complex behaviors than can evolve.
In sports the most effective deception is done automatically, without thinking.
Ah. That’s in interesting point, I think, and one that kaiboshes Libet IMO. We can intend to act unintentionally! In fact, when we acquire automated skills that’s exactly what we do – we train ourselves so that our reactions are automated, but that means that we intend that our actions be automatic!
My favorite definition of “free will” (which I think is a tautology actually – what would will – volition – be if not “free”? What would unfree will be?) is “freedom from immediacy.
But that can include deciding in advance to by unfree from immediacy. As Ulysses did, when he had himself tied to the mast.
And then the answer is ramdom. Maybe ramdom between different pre-determined answers. So we have deterministic or ramdom(stochastic) decision-makers, which are the intlligent decision makers?
Neither, by William’s definition of “intelligent”. Both by Dembski’s.
I don’t think whether a decision-maker is stochastic rather than deterministic is terribly critical. What’s of interest is whether the output depends on the input.
I think that what`s of interest is how a stochastic or deterministic decision-maker can become able to model a distal goal.
What about the capacity to learn and to solve problems (which can be operationally defined). Possibly also the ability to use language.
Yes indeed.
Here is an interesting manual for Dr Nim. It contains a number of versions of the game plus detailed analysis of how the game works.
The interesting thing is that with only four bits of “information,” Dr. Nim can beat naive human opponents.
Is four bits within the probability bound?
I loved those instructions! It was such an awesome toy.
Four bits beats two bits. 🙂
So, in your opinion, animal mimicry is not a form of deception?