The central worry about human-Aldebaranian communication is that their biology, culture, geography, developmental history, etc, will be radically different from ours, and this presents an obstacle to our ever developing an understanding of their language. How could we begin to learn the language of a society with which we share no common ground?
A popular response to this problem is what Kukla calls the "one world, one science argument" (OWOS). Whatever the differences between humans and Aldebaranians in culture, geography, biology, morality, etc, we both inhabit the same universe, and hence the basic laws of physics, chemistry and mathematics will be the same in both places. Hence, we can expect that human science and Aldebaranian science will have the same or at least similar content with respect to these areas. Kukla quotes Carl Sagan:
[H]ow could we possibly decode such a[n extraterrestrial] message? [...] The message will be based upon commonalities between the transmitting and receiving civilizations. Those commonalities are [...] what we truly share in common - the universe around us, science and mathematics.To lay out the argument explicitly:
(1) OW: The laws of nature are the same everywhere. So:
(2) OS: Human science and Aldebaranian science will share at least some content, i.e. human science and Aldebaranian science will overlap.
(3) Finally, we conclude that this overlap will provide a common ground that will make communication possible.
Kukla first considers a couple of responses to OWOS that he doesn't develop. First, he notes that one response is to deny OW. In fact, humans and Aldebaranians don't share the same world; the laws of nature aren't the same everywhere. This response he attributes to those who hold extreme social constructivist views of science. Like Kukla, I don't find extreme social constructivism remotely plausible, though contra Kukla, it's not clear to me that social constructivists actually do deny OW. A more charitable interpretation of social constructivism is that it's just another way of denying OS.
Second, we can block the inference from OW to OS by appealing to conceptual relativism. As Kukla says, our science "depends not only on the nature of the mind-independent world, but also on the concepts and categories we employ for talking about the world." Notice that this is not to deny, as the social constructivists do, that science reveals facts about the world. Rather the point is that there may be many legitimate ways of describing the world, of "carving nature at its joints". The debate concerning different species concepts in biology is a good example of this; see my videos on the species problem (first; second). If the Aldebaranians don't share our concepts and categories, there's no reason why their science would be the same as ours, even if their science is also accurate. And why should we expect the Aldebaranians to have developed the same concepts and categories as us?
I'm not sure that this is really much of a problem. Other cultures conceptualize the world differently, yet this poses no significant threat to communication. See for instance Nisbett & Miyamoto's "The Influence of Culture": Western and Asian cultures perceive the world differently, with Westerners tending to perceive the world in an "atomistic" manner, viewing it as containing independent objects that are detachable from their context; and Asians perceiving it in a "holistic" manner, viewing objects as embedded in wider contexts and relationships. There is significant conceptual change over history; the ancient Greeks viewed the world in very different terms to us. We even find different conceptual schemes emerging in one and the same science: the example of different species concepts in biology has already been noted. In none of these cases do conceptual differences pose any problems to communicate. In general, humans are remarkably capable of adopting different conceptual schemes.
Of course, Kukla may object that for all we know, the concepts and categories possessed by the Aldebaranians would be radically different to our own, different to the extent that their overall science would be radically different - but then why should we believe that such radically different concepts would correctly describe the world? It's one thing to claim that two different conceptual schemes could both be accurate. Reflection on the many species concepts in biology confirms this. That two radically different conceptual schemes could both be accurate is a far stronger claim, and requires positive argumentation that Kukla doesn't provide.
There's another way to block the inference from OW to OS that Kukla doesn't discuss but is worth mentioning. At the very least, the move from OW to OS seems to require scientific realism: basically, the view that science correctly describes the world. We've already mentioned social constructivism, a particularly controversial alternative to realism. There are plenty of other alternatives.
Let's turn to Kukla's own arguments. He offers two arguments against OWOS: the selection problem and the superfluity problem. Most of his paper is concerned with the former.
The selection problem
Even granting OW, and granting that the Aldebaranians would have a similar conceptual scheme, Kukla thinks that the inference to OS fails. The problem is that neither of us will know all the scientific truths, and the scientific truths that we possess may not overlap with the scientific truths the Aldebaranians possess. Kukla dubs this the selection problem.
Is it acceptable to infer from OW to OS? Consider some proposition X that, if true, would falsify OS. At this point we have a mere logical possibility. (This of course means that OW doesn't logically entail OS, but nobody denies that.) The problem arises when we ask what probability to assign to X. The probability of X and the probability of OS are constrained with respect to each other. Putting it formally:
p(X) + p(OS) ≤ 1
The probability of X plus the probability of OS must be less than or equal to 1. It can't be the case that X is very likely and that OS is very likely. This point should be quite intuitive. X and OS are incompatible: if one is true, the other can't be true. So to say that X is very likely entails that OS is very unlikely. (Note: if X is false, this doesn't entail that OS is true. X's falsity simply places no constraints on the probability of OS. Even if X is false, OS might also be false for some other reason.)
X is the hypothesis that there is no overlap (or at least very, very little overlap) between human science and Aldebaranian science. If our sciences don't overlap, OS is false. So if X is likely, OS is unlikely. What, then, is the probability of X? Kukla suggests that we simply don't know. We have no good reasons for supposing that X is likely, and no good reasons for supposing that X is unlikely.
It follows that we also can't assign a probability to OS. Remember, p(X) + p(OS) ≤ 1. So if we assign a probability of e.g. 0.8 to OS, the probability of X can't be greater than 0.2. To be agnostic towards X means that we must be agnostic towards OS. And this means that the OWOS argument must be rejected. Now as Kukla points out, this isn't a decisive refutation. New evidence may arise that shows the probability of X to be low. This would put OS back on the rails. (Of course, new evidence might also show that the probability of X is high, which would refute OS decisively.) But it does show us that at present, we can't accept OS.
To summarize: Kukla's basic strategy against OWOS is to cite a proposition X that is inconsistent with OS and has an unknown probability. Since we must be agnostic about X's probability, and since X is incompatible with OS, we can't assign a high probability to OS (if OS has a high probability, this entails that X has a low probability, contradicting our agnosticism). Kukla's candidate for X is the proposition that human science and Aldebaranian science don't overlap.
Should we be agnostic about the probability of X? The defender of the OWOS argument will object that in fact, we have good reason to suppose that X has low probability. Kukla considers a number of such objections.
The fundamental laws solution: While we can't know all scientific truths, there are far fewer fundamental laws. So we can expect overlap on the fundamental laws. Kukla raises a number of objections: (A) Why should we believe there are fundamental laws? There might be no fundamental laws, but instead an infinite regress of laws, each explaining the other. A similar point here is that many philosophers tend to assume that there's a "bottom level" of reality, but perhaps it just goes on forever. Indeed, that would seem to be what induction on the past history of physics suggests; to quote Schaffer:
The history of science is a history of finding ever-deeper structure. We have gone from "the elements" to "the atoms" (etymology is revealing), to the subatomic electrons, protons, and neutrons, and now we are sometimes promised that these entities are really strings, while some hypothesize that the quarks are built out of preons (in order to explain why quarks come in families). Should one not expect the future to be like the past?(B) Even if there are fundamental laws, there might be a huge number of them. (C) There's no reason to believe that we have attained the fundamental laws. Why assume that the Aldebaranians have?
I agree with Kukla that the fundamental laws argument doesn't work, but there's a very similar argument that he doesn't discuess and that I think has more potential. The argument rests on two claims. First, Aldebaranian science must, at the very least, deal with objects and processes on a scale that's also dealt with in our science. This follows from the fact that modern physics is applied at all scales: it deals not just with medium-sized objects, but also with the extremely large to the vanishingly small. Second, there is an objective structure to the world and science tracks this objective structure. This is a statement of a limited kind of scientific realism. The point here is that even if the goal of science isn't truth, in order for science to be successful enough to make accurate predictions and form the basis for technological development, in must in some ways latch onto how the world really is. (Aristotelian physics may have been wrong in many ways, but it wasn't completely wrong, and some of the claims of Aristotelian physics are preserved, or at least "imaged", in modern physics.) If these two claims are correct, we can expect some overlap between human and Aldebaranian science. This of course is only a sketch of an argument, but it provides a suggestion for how we might justify assigning a low probability to X.
The mathematics solution: Instead of appealing to science, the defender of OWOS might appeal to mathematics. We can expect humans and Aldebaranians to share mathematics, or at least certain parts of it. Elementary arithmetic, for instance, would surely be shared between two intelligent civilizations. However, Kukla objects that the selection problems applies to mathematics as well: there's no reason to think that there's any overlap in our mathematics. Even if we all have some mathematics, there's no reason to believe that we have the same mathematics.
I don't find Kukla's objection here very plausible. Bear in mind that humans have developed a great variety of mathematical systems, including extremely abstract ones and even inconsistent ones. We clearly have a very great flexibility with regard to the kinds of formal systems that we're able to understand and manipulate. Is it really plausible that the Aldebaranians use a form of mathematics that we could get no handle on? Indeed, that we could understand none of their mathematics and that they could understand none of our mathematics? Kukla thinks we must be agnostic about this since there's no evidence either way. But this strikes me as a fairly extraordinary claim, and hence one that demands some positive evidence. In fact, I think there might be some clear evidence against it. After all, any mathematics must involve patterns. We know how to distinguish between random noise and patterns, and we're also very good at decoding patterns.
The radio solution: If we can communicate with the Aldebaranians via radio, which appears to be the only option, they must at least have the ability to construct and use radio transceivers. We must share technology. Hence we must share some science. Kukla raises two objections to this: (A) He points out that no knowledge is required in order to develop radio communication; indeed, no intelligence whatsoever is required: organisms that perceive and emit radio transmissions could arise by natural selection. (B) More importantly, there's no reason to suppose that the Aldebaranians use radio.
The common conditions solution: This solution appeals to two hypotheses. First, "adaptive pressures instil a propensity to adopt certain ideas." In other words, natural selection has built in us innate tendencies to adopt certain beliefs about the world. If green berries are poisonous to us, it will promote survival and reproduction for us to be wary of green berries, for us to think that eating green berries is dangerous. Second, "there are conditions for survival that can be expected to obtain wherever intelligent life evolves." Kukla doesn't accept either of these hyptheses, but is prepared to grant them for the sake of argument (I will return to the latter hypothesis later). The defender of OWOS then argues as follows: Since both we and the Aldebaranians evolved in common conditions, we are likely to have faced similar evolutionary challenges. These similar challanges would have lead to similar solutions. In some cases, these solutions will take the form of certain true beliefs about the world. So, there are certain truths that both humans and Aldebaranians are likely to be innately disposed to accept. These truths are likely to be used in both human and Aldebaranian science.
This strikes me as an extremely silly, wildly speculative argument, leaping from premise to premise without any adequate justification. The appeal to common conditions can, I think, furnish a defense of the plausibility of human-Aldebaranian communication (again, I'll return to this later), but this particular defence is obviously inadequate. I won't deal with Kukla's discussion since I agree with him that argument doesn't work, and some of the problems are simply obvious. First, why believe that we face similar evolutionary challenges? Of course, at a very abstract level, all life forms must face similar challenges: how to find food, how to stay alive, etc. But at a finer grain, even species that evolve on in the same areas on the same planet don't share the same challenges; the lion doesn't face the same evolutionary challenges as the capybara. Second, why believe that similar challenges lead to similar solutions? Third, even if the argument is right that humans and Aldebaranians are innately disposed to accept the same beliefs about certain things, it's worth noting that the history of science is one of a constant violation of our intuitive beliefs. There's no reason to think that intuitive beliefs are likely to be preserved in scientific theories.
Finally, Kukla gives a transcendental argument for the selection problem: he argues that there's good reason to believe that the selection problem can't be solved; we can't ever show that X has a low probability. The reason is that we have no good reason to believe that the Aldebaranians have any science or mathematics whatsoever; and if they have no science or mathematics, a fortiori our science and mathematics have no overlap with theirs. All we know about the Aldebaranians is that they're intelligent beings. Intelligent beings need not have developed science and mathematics. Humans have always been intelligent, but we didn't develop science and mathematics until relatively recently.
Kukla is of course right about this. But I think defenders of the OWOS argument could argue that he's simply changing the subject. Is the only thing we know about the Aldebaranians that they're intelligent beings? No: we also know (because we've stipulated) that they've reached a high level of technological development. (Recall the passage from Sagan that Kukla quoted. We are to imagine a civilization that can transmit and receive interstellar messages. Of course, Kukla will point out, re his response to the "radio solution", that organisms need not have any technology or even intelligence to develop radio communication. While he's right that the basic elements of radio communication could arise by natural selection, reception and transmission of interstellar signals is another matter.) Surely nobody who's proposed the OWOS argument has in mind communication with extremely primitive extraterrestrials. Obviously "one world" doesn't entail "one science" if the extraterrestrials in question have no science whatsoever, just as "one world" doesn't entail "one science" if the extraterrestrials in question are merely simple microbes.
The superfluity of OWOS
Kukla's second objection to OWOS is that it's superfluous. Recall that the point of OWOS is to explain how we would be able to communicate with the Aldebaranians despite the apparent lack of any "common ground" between us. OWOS is supposed to show that actually, we do have common ground: namely in our basic science and mathematics. However, Kukla points out that we don't need to share any common ground in order to learn another person's language: "Babies do it all the time: they learn the language of beings with whom they share no prior topics of discussion, and they do it in a few years' time without expending any extraordinary effort." The correct response to the communication problem is not to demonstrate that we would share common ground with the Aldebaranians, but instead to reject that any common ground is required.
In a sense, of course, Kukla is right that babies don't share any "topics of discussion" with adults, just in virtue of the fact that babies aren't capable of discussing anything with adults. This however is completely trivial. In the same sense, the man who can only speak English shares no topics of discussion with a person who can only speak Japanese. So Kukla is just equivocating on the notion of "common ground". In the sense of "common ground" that people debating the OWOS argument are interested in - the sense that allows me to say that the English speaker shares a lot of common ground withh the Japanese speaker, but possibly no common ground with the Aldebaranian speaker - babies share substantial common ground with adults. They live in the same culture, they will experience the same geography, they have very similar biology, etc. They may even share a kind of innate "language module", as Chomsky has suggested with his theory of language acquisition.
Ultimately then, I think that Kukla's objections to OWOS are inconclusive. Before ending this post, I'd like to note a couple of points about the debate. First, one notable assumption by all parties in this debate is that communication would have to be linguistic. But why couldn't we, for instance, use pictures?
Second, defenders of OWOS grant that the only thing we could have in common is science and mathematics. I think this concedes too much to their opponents. Humans and Aldebaranians will probably have much more in common; I list three points. First, on any world, objects will have to move in much the same ways as they move on this one. The basic facts of mechanics and gravity are going to be same, no matter what world you live on. Second, there are strict limits on the kinds of chemistry that could be the basis for life, especially any life that could reach the complexity required for intelligence. Third, we can expect that humans and Aldebaranians will exhibit similar behaviours on a very abstract level. The Aldebaranians must have developed the capacity for experiencing something like pain, for instance, which will evince itself in withdrawal from harmful stimuli. Kukla is wrong to reject that we can expect common conditions for humans and Aldebaranians.
This is the real basis for the superfluity of the OWOS argument. In order to have good reason to believe that there's significant overlap between human science and Aldebaranian science, we must believe that there's some sort of overlap in the conditions in which we live. But this already provides the common ground needed for communication.
No comments:
Post a Comment