Saving the symmetry principle, IIIb: truth in the history of science

This post continues my effort to understand the symmetry principle by distinguishing different senses of the claim “people do not believe things because they are true.” As you can see, this is not an easy job: this post adds 5 readings to the 6 discussed in my previous post. But nor is it an exercise in hair-splitting or nit-picking. I'm not suggesting that we need to make these distinctions explicit whenever we discuss the symmetry principle, the nature of scientific truth, or the role of evidence in settling scientific debates. But our discussions of all those topics would be improved if we kept these distinctions in mind when we formulate our claims and when we assess the claims of non-historians. (Readers who are pressed for time may want to skip to the end of this post, where I summarise my 11 readings and draw some morals from them.)

7. The truth-value of any given theory is obvious once you decide to consider the evidence. Call this the “self-evidence assumption.” Most historians of science think that this is a dangerous assumption, and that it underlies much popular writing about science. Vanessa alerted me to the relevance of this assumption when she wrote in a tweet that “symmetry is a starting point for undermining the 'if they'd've looked, they'd've believed' assumption.”

I agree that this is a dangerous assumption, and also that it is relevant to the symmetry principle. However I would say that the wrongness off the assumption is the starting point for the symmetry principle, rather than the other way round. This is important: the main reason people give misleadingly asymmetrical accounts of past scientific debates is that they underestimate the quantity, variety and complexity of the evidence that lies behind any well-grounded belief about nature.

This underestimate encourages two other false assumptions. The first is that the evidence available in the past for any given theory was more or less the same as the evidence available today. If you think that the theory of evolution from natural selection follows from a few simple inferences from everyday observations, you are unlikely to appreciate the fact that Darwin had less evidence for that theory than we do today. And if you do not appreciate that, you are unlikely to appreciate the reasonableness of Darwin's nineteenth-century opponents.

The other consequence of the self-evidence assumption is the belief is that, once there is strong evidence for one theory over another, it is impossible to mount a plausible case for the rejected theory. If you think that the evidence for evolution is blatently obvious, you are likely to think that its present-day opponents are stupid, biased, or insincere.

In fact it is possible to make reasonable objections to just about any piece of evidence for evolution available today. I believe that the vast majority of these objections can be answered (otherwise I would not believe in evolution). However I also think that a full answer to those objections would require specialist knowledge of biology and paleontology, not to mention genetics, biogeography, molecular biology, anatomy, and maybe some philosophy of science. This means that it is possible for non-specialists to build fairly plausible cases against evolution by natural selection.

Does the falsity of the self-evidence assumption mean that evidence and argument play little role in scientific disputes? Of course not. If anything, it shows that there is more evidence, on both sides of any debate, that we might first imagine.

8. If a statement is true, it corresponds to reality. Rebekah Higgitt responded to my first post in this series by tweeting that “I do see statements that scientific theories/facts are true because they're true (ie true reflection of reality).” My response was that there is nothing wrong with this, if people mean just that truth consists in some sort of relation between a statement and the world. This is roughly what philosophers call the “correspondence theory of truth.”

I sometimes come across the view in science studies literature that the correspondence theory is a hopelessly naïve theory of truth that was abandoned by all right-thinking people some time between 1950 and 1990. Granted, philosophers continue to argue about whether or not the correspondence theory of truth is a good one, and about which correspondence theory is the best one—witness this article. But the fact that Stanford Encyclopaedia of Philosophy has an up-to-date article defending the theory suggests that it is far from a minority view.

9. The evidence for a theory is a good guide to the truth-value of the theory. According to this view, we should believe theories to the extent that they have good evidence in their favour. This may seem like commonsense—after all, if the evidence does not tell scientists what to believe, what does?

Nevertheless this view is often rejected by both scientists and philosophers of science. For example, IanLove commented as follows in reply to one of Vanessa's posts: “In science saying you believe something means that you consider it best fits the evidence..... As for truth - that is best not used: because, as others have said, all scientific theory and evidence is work in progress.”

I find it hard to understand this blanket quietism about truth. Why would scientists collect all that evidence for their theories if they did not think it would lead them closer to the truth? True, many philosophers of science deny that evidence is a good guide to the truth-value of every kind of belief. But even those philosophers (such as Bas Van Fraassen) usually say that scientist's beliefs about observable phenomena are probably true, and that the evidence is a good guide to the truth-value of that kind of belief.

But suppose for the sake of argument that IanLove is right, and that scientists endorse beliefs that best fit the evidence but never take the further step of saying that those beliefs are true. This would mean that the evidence can never explain why scientists believe theories to be true (since, by hypothesis, they never believe that of theories). But for the same reason, social or institutional factors could never explain why scientists believe theories to be true. So even IanLove's anti-truth stance is no grounds for preferring social or institutional explanations over evidential ones.

10. Historians can explain the past development of things that are defined in present-day terms. I include this one because it came up in a comment on one of Vanessa's earlier posts, the one that prompted Vanessa's post on the symmetry principle. The post was about how we might explain the decline in incidence of TB in the twentieth century—was it drugs and vaccines, or improved nutrition, or perhaps public health measures like clean water and better sanitation?

Vanessa noted that the meaning of “TB” has changed over time, and that this causes problems for any attempt to explain the change in its incidence over time. This provoked the following comment from Wolfbone: “All you have to do is pick the modern, most informed, definition of what TB is and do your research and write your history in the light of that knowledge.” Vanessa subsequently presented this in a tweet as a clear example of bad historical practice, presumably because Wolfbone was proposing that we think about the past in present-day terms.

As I have said elsewhere, I do not see what the problem is with thinking about the past in present-day terms. Sure, you are going to miss a lot if you ignore earlier, different definitions of TB in your history of the topic. But you are also going to miss a lot if you fail to adopt a consistent definition of TB. In particular, you are going to miss the opportunity to explain why rates of TB incidence changed over long periods of time (rather than just giving a sequence of disconnected explanations of how various TB-related conditions changed during the short periods in which each one of those conditions was thought to define TB).

I can see why a historian might choose either one of those approaches, but I do not see why one would want to eject the present-centred one from the canons of good history. As Wolfbone put it, “You can perfectly well write a history of TB and a history of “TB.””

Vanessa's worry seemed to be that the present-centred approach is “progressive” in the sense that “it starts with the assumption that we're obviously right now, and were therefore obviously wrong then.” This statement is imprecise in just the place where precision is needed: the starting assumption is that today's theory is considerably more likely to be true than yesterday's theory, not that today's theory is "obviously right" (if that phrase means "certainly right" or "obviously right once you decide look at the evidence").

And even if the assumption were that we are “obviously right” now, this would not commit us to the belief that we were “obviously wrong” in the past. Perhaps we have uncovered some new evidence recently that means that the truth of our current theory is much more obvious now than it was a decade ago (see 6. above).

11. Today's theories are more likely to be true than yesterday's. Wolfbone suggested that the real reason for Vanessa's hostility to the use of the present-day definition of TB was her assumption that the present-day definition is no better—in the sense of being no more well-supported by the evidence—than previous definitions.

Wolfbone wrote that Vanessa was “apparently motivated by the false belief that today's science facts are just as fragile as yesterday's.” And indeed, Vanessa wrote (for example) that “diagnosis and disease definitions change all the time; today's is as likely to be proved 'wrong' as yesterday's.”

So, does the fact that we have been wrong in the past mean that today's theories are no better than yesterday's? Philosophers of science have long pondered this argument. There is even a name for it: the “pessimistic induction”. The debate is complicated, with plausible arguments on both sides. This means that it would be unwise for historians to base the central tenet of their field—the symmetry principle—on the presumed outcome of the debate.

Do historians need to take this risk? That is, does the symmetry principle stand or fall with the pessimistic induction? In one sense the answer is “no.” Imagine that we were absolutely certain that the shading of the moon is due to its surface relief rather than the unenven density of its internal matter. Never mind how we might have become certain, or whether we are actually certain—just pretend that we are. Now, this certainty would be perfectly consistent with the belief that Galileo's evidence for the rockiness of the moon was no better than the Jesuit's evidence for the uneven density of the moon.

But there is another sense in which the symmetry principle does appear to stand or fall with the pessimistic induction. How would we become certain that Galileo was right? Presumably by looking at the best available evidence. But to make this inference, we must suppose that now (May 2013) there is a close connection between the truth-value of a belief and the state of the evidence. But there is nothing special about May 2013, so the same connection must have existed in the 1600s, when Galileo was up against the Jesuits. But this claim is inconsistent with the symmetry principle, which denies that the truth-value of Galileo's belief is any guide to the evidence available to him. So it looks like we can have the symmetry principle, or believe present-day scientific theories, but not both. Maybe Vanessa was right to hitch the symmetry principle to the pessimistic induction.

Although I think that linkage is mistaken, that is not my point here. My point is that we need to break that link in order to save the symmetry principle. Faced with a choice between the symmetry principe and trusting present-day science, many people would ditch the symmetry principle. And that would be a perfectly reasonable choice. Granted, to historians it is absurd to say that all past scientists who held true beliefs did so on the basis of good evidence, whereas those who got it wrong did so on poor evidence. But it is just as absurd, if not more so, to say that today's science is no more likely to be true than yesterday's.


Conclusions

There are lots of legitimate ways in which questions of truth and falsity can enter into historical research. The fact that something is the case can help to explain why people believe it to be the case (#3 in the previous post). The fact that something is the case about nature can also explain why people believe something else to be the case about nature (#4). People can believe things partly because of the evidence (#1), including the factual evidence (#5). And historians can legitimately explain the past development of things (like TB) that are defined in present-day terms (#10 in this post).

In my view, there is little room for debate about these issues (except perhaps #10). They should be distinguished from other, deeper issues that are the subject of ongoing debate among honest and well-informed philosophers. Does truth consist in the correspondence between a statement and reality (#8)? Is the evidence for a theory a good guide to the truth-value of the theory (#9)? Are today's theories more likely to be true than yesterday's (#11)?

Historians should not blithely assume that the answers to these questions are all “yes.” But nor should they assert that the answers are “no” and then build their historiographical principles on this assertion. It is undesirable, and probably unnecessary, for historians to base their methods on claims that are controversial among mainstream philosophers.

There are other forms of historical explanation that do make illegitimate use of truth or evidence. People do not believe things to be true because they believe them to be true (#2). Nor do they respond well to dogmatism (#6). And above all, it is not the case that the truth-value of any given theory is obvious once you decide to consider the evidence (#7).

However, the fact that these kinds of explanation are illegitimate does not threaten the claim with which I began my previous post, viz. that evidence and argument are on a par with social, political or institutional factors when it comes to explaining the beliefs of scientists and laypeople, whether in the past or the present.

The main aim of these two posts was to clarify the symmetry principle. It should now be clear that the principle is not a blanket ban on the invocation of truth, evidence or reality in historical writing. Instead it is a ban on a rather special way of using those concepts: it is a ban on inferences from the truth-value of a past scientist's belief to a decision about whether to explain that belief in terms of evidence and argument or in terms of something else. In other words, it is a ban on what I called “The Fallacy” in my first post in this series. But it is no more than that. In particular, it has very little to do with most of my 11 readings of the claim that “people believe things because they are true.”

In the next post in this series I will describe another way in which historians exaggerate the scope of the symmetry principle. But first, a political interlude. Expand post.

Malheureusement je ne traduis plus tous les posts que j'écrit sur ce blog. Si vous voudriez lire les posts qui ont été traduits, veuillez cliquer ici.

Agrandir ce message.