Never Trust Someone Who Worries About Computers Behaving Badly (as Opposed to Malfunctioning)

This is an absurd article:

EXCLUSIVE: Dr David Levy told Daily Star Online it will be important to teach the robots of the future about consent, because they will have their own sexual desires

Presumably they mean this David Levy, who has a PhD in some unrevealed subject, and he is promoting an old book based on his dissertation called Love and Sex With Robots. If this got him a PhD in computer science, some computer science department needs to have its charter reviewed, because he reveals a shocking level of ignorance about how AI works.

Levy presents an image of AI where the computer has its own drives, and like a human is able to find novel solutions to satisfying those drives–neither of those things is true. Computers do not have drives. You don’t program a computer to want to calculate an orbit to the moon to make the computer do that, and you don’t program a computer to want to simulate sex to get the computer to do that.

The computer is not feeling anything like pleasure or the satisfaction of a drive; it is just adjusting the position of a collection of stepper motors in response to the input of various sensors, and it can only do that to the extent that it has been programmed to make the right adjustments based on the right inputs. A robot has no capacity even to subdue someone unless you have programmed that in. A robot that has not been programmed to hold someone down cannot hold a woman down and rape her.

There are some caveats to my claim that computers can’t come up with novel solutions. Machine learning is a technique where you program the computer to try random actions and record the outcomes so it “learns” what those actions will do without a programmer specifically having to program that “knowledge” in. This is a very helpful programming technique when the actions and results happen at computer speeds because it saves a lot of time; the computer can try millions of actions in the time it would take a programmer to program just one. But the computer has no ability to judge what to try, and it doesn’t even have the ability to try anything that hasn’t been programmed in. In particular, if the programmer has not programmed the computer to try restraining someone, it will never try that.

And it isn’t likely machine learning could be used for something like teaching a robot to interact physically with humans, because it can’t try millions of things per minute. Imagine getting a robot to learn how to simulate sex. You would have to program in all of its potential actions, which would involve moving body parts by a fraction of an inch at a time. A quick perusal (by me, I haven’t looked it up) suggests that with multiple degrees of freedom for many joints, there are about 36 dimensions just in the major joints (not counting fingers and toes). You have to try all combinations of motions, both up and down, so at any point, there are at least 2^37  or 137,438,953,472 different possible motions to try–and each motion only moves body parts by a fraction of an inch.

Adding to the difficulty, you are trying to get the robot to please a human sexually, so there has to be a human involved in the testing to indicate whether the robot is successful or not. This is not a viable technique, so sex robots will not be programmed using machine learning–at least that won’t be the primary technique, and no robot will ever learn to subdue a human by machine learning–it would have to be explicitly and deliberately programmed in.

Now, it’s probably possible to program that in. A robot could potentially be programmed to restrain someone while simulating sex with them, but that would be deliberate programming, and there is nothing to prevent it that would correspond in any way to making the robot understand consent.

Another possibility is that a robot could possibly be programmed both to subdue an intruder and to simulate sex, and there could be bugs where it would combine these two functions with a result that the robot would involuntarily restrain someone and then simulate sex with them. But the solution to that potential problem is not to program the robot about “consent” either; the solution is some redundant lockout mechanism in the software that prevents both restraint mode and sex-simulation mode from being active at the same time.

The entire notion of programming a robot to understand consent is ridiculous. You can’t get a robot to genuinely understand anything. They respond to their sensors in ways that they are programmed to respond. Could you program a robot to sense a person’s behavior and respond in ways that would make the robot change its actions based on that person’s behavior? Probably, but describing that as making the robot understand “consent” is absurd.

Is Pain Evil? Are Moral Truths Describable in Non-Moral Language?

This post asks two questions. The first, “Is pain evil?” has been widely discussed in the past, but I think it’s worth bringing it up again with regards to the One-Chip Challenge (see below). The second, “Are moral truths describable in non-moral language?” is something that I haven’t seen discussed elsewhere (full disclosure though, my reading in moral philosophy is not extensive). Notice that I’m not asking whether ethical facts, are reducible to non-ethical facts, but whether it is possible, even in principle, to state rules using non-moral language which accurately and completely describes moral truth.

A Vertiginous Revelation

This post at Instapundit made me realize something a bit dizzying. Both Donald Trump and Mitt Romney had me fooled. I thought Trump was a pathetic attention-whore with no moral foundation, and Romney was a man of character who could be relied on to keep his word. It turns out that Romney is a pathetic attention-whore with no moral foundation and Trump is a man of character who can be relied on to keep his word.

There is No Estimate of Pi in the Bible

There is no estimate of the mathematical constant pi (or ) in the Bible. Mathematically-inclined critics sometimes quote I Kings 7:23

And he made a molten sea, ten cubits from the one brim to the other; it was round all about, and his height was five cubits, and a line of thirty cubits compassed it round about.

and point out that since the numbers here are the diamater and ratio of a circle, this implies that whereas is actually a bit more than 3; it’s actually a bit more than 3.14159 which is as far as I can go by memory. After this, they typically add some snarky remark about Biblical accuracy, but the criticism is wrong-headed for two different reasons, either one of which is sufficient to make anyone who spends a second thinking about it realize that the criticism is dumb.

First, this is obviously not a mathematical estimate of . There is no indication here or anywhere else in the literature of the region and period that they even knew that the ratio of diameter to circumference of a circle is a constant; that is, there is no evidence that they knew there was a constant to estimate. What this passage clearly reports is not a mathematical theory, but a pair of measurements, rounded to the nearest cubit.

The entire passage is full of measurements, clearly someone went through the construction site with a measuring line and measured a bunch of stuff while a scribe wrote it all down. The text first specifies the size of the basin by its diameter, implying that they measured the diameter, and then specifically says that they measured the circumference with a line. If you measure the diameter of any circle and it rounds to ten units, then measure the circumference with a measuring line and round to the nearest unit, the most likely result to get is 30 units. That is not an error; it is a rounded value.

Second, people who claim Biblical inerrancy are not claiming high-precision mathematics. If the writer of this report been estimating (which he wasn’t) then his estimate was accurate to within the precision of the report. There is no way to actually write down all of the digits of because there are an infinite number of them. You have to stop somewhere, and there are lots of applications where stopping with 3 works just fine. In other words, even if this were an estimate of , it is not an error; it is just less precise than the critic would like.

I’m not a Biblical literalist, but horrifically bad arguments like that one just annoy me because they reveal that the critic doesn’t really care about the truth, but only about a gotcha.

A Short Note Observing that Physicalism is a Rationalist Doctrine

By “physicalism” I mean that doctrine which says that only the physical or material world exists and that all causes are physical causes. By “rationalism” I mean that doctrine which says that we have access to information about the physical world which does not come from our physical senses alone. By “empiricism” I mean the denial of rationalism.

Physicalism and empiricism are often thought to go together, but in fact a consistent empiricist must deny that there is any evidence for physicalism. To see this, we note that the empiricist is committed to the position that all of our knowledge comes from our physical senses; there is a sort of wall of separation between our minds and the physical world such that all information about the physical world must come through a fixed and limited sort of pipeline of information, namely the senses.

Our epistemological situation as envisioned by the empiricist can be modeled as follows: a man sits in a room which has a slot in the wall. A paper tape issues from the wall, and on that tape is a sequence of letters. As time goes on, more of the tape issues from the wall with more letters. The man has no information about where the tape comes from or how the letters are being produced.

Suppose now that by examining the tape, the man is able to discover a pattern. Say that a sequence of three A’s is always followed by a B. Clearly the man can use discovered patterns to predict future letters, so long as the pattern continues to be followed, but what can the man infer about the causality that leads to the pattern? He has no idea how the pattern is being produced; it could be by another person, by a computer (in which case he has no knowledge about the algorithm), by some sort of mechanical device, or anything else.

It would be reckless for the man to come to any conclusions about the cause behind the pattern, but it would be more than reckless–it would be downright bizarre–to say that the letters themselves cause the pattern. It would be odd to say that the A’s combine their forces in some way to produce a B.

This bizarre position is the position of an empiricist physicalist. The empiricist can see nothing but the letter, which corresponds to a physical phenomenon; the mechanism behind the wall–the noumenon– is entirely unavailable. As a physicalist, he concludes that the letters themselves–the phenomena–are the causes. This is a great leap of faith, and a bizarre one, for an empiricist to take since nothing in his observation, pared of all intuition and expectation, can possibly tell him that phenomena have causal power of any sort. All his senses tell him is that phenomena exist and follow patterns.

The only way to justify the idea that phenomena have causal power is to rely on physical intuition. When we take a stick and push a ball, we have an immediate sense that it was our hand that made the stick move and the stick that made the ball move, but relying on such intuition is a rationalist position, not an empiricist position.

So, just the position that some phenomena are caused by other phenomena, is a rationalist position, must less full-on physicalism which asserts that all phenomena are caused by other phenomena. Even more a violation of empiricism is the idea that there is nothing behind the wall at all. How could an empiricist possibly know? Consequently, anyone who is a physicalist can’t be considered an empiricist.

Is it a Substantive Question Whether Time is Static or Dynamic?

Bill Vallicella  writes:

“Surely it is a substantive question whether concrete, mind-independent reality is static or dynamic”

I can’t imagine how the question could be substantive. I take the meaning of “substantive” to be that the universe would be a different place in one version of reality than in another, and I can’t see how that would be.

I believe from the context that what Bill is referring to is these two theories of time:

• The dynamic theory of time, or the A-theory posits that the world is a place with distinct past, present, and future, such that the present is continuously becoming the past.
• The static theory of time, or the B-theory posits that time is an undifferentiated dimension, in which past, present and future are just how we perceive the universe, much like as we travel along a road, we perceive the road behind, the road where we are, and the road ahead.

If the choice between these two theories is substantive, then it must be the case that there would be a real difference in the universe if the A-theory were true versus if the B-theory were true. Is there such a difference?

Well, if the A-theory is true, then the B-theory is also true in the sense that it is an abstraction of A-theory time, where you abstract away the distinction between past, present, and future, viewing the whole of time as an undifferentiated line. Conversely, if the B-theory is true, then the A-theory is also true in the sense that it is an application of the B-theory, where each mind represents the past, present, and future of each point based on where the point occurs in the line of time.

How would the universe be different if the A-theory were true and the B-theory an abstraction vs. the B-theory being true and the A-theory an application? Well, we might say that if the B-theory were true, then things in the past, present, and future all exist equally, while if the A-theory were true, then only present objects would exist. That won’t do, because it doesn’t represent a difference in fact, but only a difference in terminology. Both the A-theorist and the B-theorist agree about the facts of the matter, they merely disagree on how to speak of them. The A-theorist agrees that past objects did exist and that future objects will exist; he just emphasizes that such objects do not presently exist.

Meanwhile the B-theorist agrees that the objects that the A-theorist calls “past” are in fact not present in the section of time that the A-theorist calls “the present”, and similarly for future objects. Consequently, there is a direct mapping from A-language to B-language and an inverse mapping from B-language to A-language, so that they are merely describing the same thing in different words. A similar argument can (I claim) be constructed for any difference in descriptions.

But Bill adds another comment that might lead us to a real difference:

Is temporal passage real or is it mind-dependent?

Is there a substantive question of whether time is mind-dependent or not? I say “no”, on the grounds that both theories of time imply that time is mind-independent. Clearly, if the A-theory is true, then the past, present, and future are mind-independent facts of the universe which the mind perceives. I’m not sure what it would mean for an A-theorist to think that time itself (as opposed to the perception of time) is mind-dependent.

If the B-theory is true, then presumably at any point t on the timeline, the mind perceives t as the present, times previous to t as the past, and times after t as the future. How can this be considered mind-dependent? It is just asserting that the mind, at each point in time, possesses a correct perception of its position on the timeline. One might counter that the mind also possesses a sense of the passage of time, which is mind-dependent, since time is not actually passing, but this is just another way of saying that the mind is aware, at each point along the timeline, that there are points in the past which the mind in the past perceived as the present.

I suspect the notion of time being mind-dependent is an artifact of A-theory minds trying to take a B-theory perspective, and inadvertently relying on an A-theory concept while doing so. In particular, I think the idea is that the mind is a sort of focus traveling along the B-theory timeline, viewing it’s current location at each time as the present. But of course, if the B-theory is true, then there is nothing moving along the timeline since the timeline is static. Such a notion requires a sort of meta-time, which reintroduces all of the problems that the B-theory was supposed to deal with.

So, I think the notion that there is any substantive difference between the A-theory and the B-theory is very dubious. Anything that an A-theorist believes about time can be translated into something that a B-theorist believes about time and vice versa, which makes the debate seem more like a terminological dispute than a substantive one.

Your Cat Doesn’t Love You

I’m sorry, but it has to be said. All of you cat people basking in the warmth of your pet’s love and adoration are suffering from a delusion. When kitty purrs in your lap she isn’t trying to communicate, “I love you”; she is trying to communicate “You do your job well, serf. I shall permit you to live another day.” When kitty leaves a dead animal on your door step, she isn’t trying to give you a gift; she isn’t motivated by love; she just got bored waiting on the doorstep for you to open the door so she dropped it and wandered off.