Book review - Everyday Practice of Science

I've been sent a review copy of Frederick Grinnell's Everyday Practice of Science: Where Intuition and Passion Meet Objectivity and Logic (Oxford University Press, 2009). Grinnell is a practising scientist (a cell biologist) and an expert in scientific and medical ethics. That's an impressive combination, so I felt very positive about the book when I opened it. It promised to do a good job of explaining how the practices of working scientists relate to the arguments that are eventually used to support scientific findings.

Grinnell makes a pretty good fist of this - although there is relatively little that would be unfamiliar to philosophers of science or to scientists who reflect on the foundations of the scientific enterprise, it's explained clearly, generally seems plausible, and gives due acknowledgment to both the messy elements of actual scientific practice and the sorts of arguments that are used to support scientific theory. There need not be an easy match - e.g. data obtained from a series of experiments meant to test hypothesis A may, serendipitously, give some help to hypothesis B that the researchers didn't even have in mind. While Popperian arguments based on hypothetico-deductive reasoning may provide powerful support for some body of theory, the actual process that was followed may have been rather different. When scientific papers are published, they may leave out much of the messiness of what actually happened, day-by-day in the lab, and instead describe the experiments in a way that reveals the logic of the argument.

By all means read this for yourself. I have no difficulty with this point, and even Karl Popper realised that something like this is true. He never claimed that hypothetico-deductive reasoning describes the actual process of scientific discovery, only that it is the logic of scientific discovery. Even that may be an exaggeration, as not all scientific arguments are necessarily about how observational data corroborate or falsify hypotheses. Still, these arguments are used, and it is useful to think about how they relate to actual practice.

I'm not sure I agree with all Grinnell's views about research ethics, but let that pass. Well, except for this: I can't help but cavil at a reference to Peter Singer's "moral skepticism", when the views discussed actually have nothing to do with moral scepticism. Singer may, in fact, be committed to a form of moral scepticism, if he's pursued far enough, but he's always been rather ambiguous about that. He is mainly known for his views about normative ethics, not metaethics, and here he does not look like a moral sceptic at all. He does reject many substantive moral positions that he considers wrong, and he is certainly sceptical about the content of traditional morality, but that's a very different point.

Moral scepticism is a metaethical position (or, rather, a cluster of such positions) from which there seems to be something bogus about the whole enterprise of morality or moralising. For a moral sceptic, there's a sense in which morality is bunk, that it can't deliver on what it claims (though there may still be perfectly good reasons for you to act kindly and cooperatively, to encourage others to do so, and so on). None of this has much to do with Singer, and certainly not with the views that Grinnell discusses.

My big problem with the book is rather different, and relates to the never-ending argument about religion and science. Grinnell considers their relationship in a chapter of almost 30 pages in a book of about 200 pages, so it's a good-sized chunk. He argues throughout the chapter that religion and science are "complementary", but his arguments for this are very weak.

The main point seems to be that science and religion cannot conflict if religion minds its place and only talks about matters of purpose, meaning, and an unseen world beyond "shared sensory space". This is, of course, a version of Stephen Jay Gould's principle of non-overlapping magisteria. And I agree that religions can avoid clashing with science if they thin out their truth-claims until they are no longer offering explanations of the way things are in the world that we perceive through the senses. That, however, is a lot to demand of religion.

But even if religion keeps to what Grinnell (like Gould) thinks is its proper turf, it's not at all clear why it is thought to have correct answers to questions about the "meaning" or "purpose" of life, or about the existence or character of any unseen order, or about moral questions. I see absolutely no reason to think that religion is authoritative on any of this, and Grinnell doesn't provide one. On the contrary, he notes that there are many different sets of supposed religious truth about the unseen order, etc. Given this "fragmentation", why assume that any of these sets of "truths" is actually correct?

Grinnell says: "Science provides the technology for doing things. Religion provides the values to do what should be done." But that is false. Some values may come from religion, but others may be innate - I may have an inborn tendency to value my own survival, or to value getting food, water, and sex - while others may be taught by parents or the larger culture, or emerge from reading literature or from a process of philosophical introspection, or from somewhere else again. To suggest that religion is the source of our values, working in a complementary way with science, is nonsense at worst and an extraordinarily controversial claim at best. Inverting a famous statement by Einstein, Grinnell says that "Science without religion is blind," but this is just not right.

There is no reason whatsoever to think that we need religion as our source of guiding values, or to prefer whichever values do come from religion to any others that we might obtain from numerous other sources. There may be good reasons for science to enter into productive relationships with literature, or the law, or secular philosophy, by why should it take any notice of religion? No reason at all that I can see, except a wish to pander to religion's grandiose self-conception as a fount of values and otherworldly knowledge.

What's probably true is that we need affective attitudes (desires, hopes, fears, values, and so on) as well as factual knowledge ... or we will never be motivated to do anything. But no one can seriously maintain that religion is or should be the source of all our affective attitudes. That would be a mad claim. How do intelligent people like Grinnell come to think this sort of stuff, and how do they get away with publishing it?

I don't want to be too hard on Grinnell in particular: he has written an interesting little book. But I'm tired of ill-evidenced claims that religion (or some particular religion) somehow provides an authoritative source of values or moral guidance or knowledge about some hypothetical "unseen world" or ... No, stop - we have no reason to think it provides any of those things. It's time to stop saying this placatory stuff about religion. Religion is something that science can well do without - and so can we all.