Tuesday, 25 June 2013

Bait-and-Switch: Arguing Against Scientific Consensus


Arguing Against Scientific Consensus

Author's Note: This is an expanded version of this comment I wrote, with reference to this comment, by a regular commenter and apologist for various forms of purported therapeutic products and services which might more accurately be described (at least in this author's opinion) as "rank quackery".

In politically-controversial (if not scientifically-controversial) bodies of knowledge with large, well-settled factual support, it is a common occurrence to encounter those arguing against those bodies of knowledge, specifically by attacking what would otherwise be the entirely unsurprising near-unanimity - or, as used in the rest of this post (and elsewhere), the consensus - among relevant subject-matter experts.

There are two rhetorical devices that can be used to mount such attacks. The first, which I shall not examine in detail here, is the technique of asserting the existence of a countervailing bloc of subject matter experts, a bloc which is asserted or insinuated to be of some significance as compared to the subject matter experts upholding more mainstream positions. Such is the technique used by the OISM Petition Project (discussed here and here at the climate science website Skeptical Science) with respect to global warming/climate science, or by, say, various forms of creationists with respect to biological evolution of organisms.

The second rhetorical device, which is the subject of this post, is to attack the very notion of scientific consensus among subject-matter experts. Perhaps the most famous such line of argument was made by the late physician, author, and film/TV producer Michael Crichton:

Historically, the claim of consensus has been the first refuge of scoundrels; it is a way to avoid debate by claiming that the matter is already settled. Whenever you hear the consensus of scientists agrees on something or other, reach for your wallet, because you're being had.

Let's be clear: the work of science has nothing whatever to do with consensus. Consensus is the business of politics. Science, on the contrary, requires only one investigator who happens to be right, which means that he or she has results that are verifiable by reference to the real world. In science consensus is irrelevant. What is relevant is reproducible results. The greatest scientists in history are great precisely because they broke with the consensus. There is no such thing as consensus science. If it's consensus, it isn't science. If it's science, it isn't consensus. Period.

It would seem, on the surface, that there is no reasonable objection to Crichton's argument. After all, it is, as far as I am aware, widely known that appeals to authority or to popularity are logical fallacies and have no place in scientific reasoning. It follows that if an appeal to a scientific consensus is an appeal to popularity, or to the authority of scientists, then it has little to no bearing on whether a set of claims, ostensibly based on scientific research, is true (up to the asymptotic limits of truth available using inductive methods and due to the fallibility of human minds).

Throw in a reference to Galileo, perhaps, and a Q.E.D for good measure, and all that consensus talk is safely dismissed as rubbish, right? Right?

Wrong. The argument against scientific consensus amounts to little more than a bait-and-switch. What follows is how and why.

Breaking It Down

To make this analysis a little easier, let's try and set up a reasonable generic argument against consensus, in a point-by-point argument form, using Michael Crichton's argument as a template:

Premise #1: An appeal to authority and/or popularity is illegitimate in scientific discussion.
Premise #2: Arguing that a scientific claim is correct on the basis of the existence of a consensus by subject-matter experts amounts to an appeal to authority and/or popularity.
Premise #3: Historically, notable scientists have overturned or broken down existing scientific consensus - and it is this overturning which has made them notable.
Conclusion: The argument from consensus is illegitimate.

With that done, let us now analyse each step in this argument.

Premise #1

An appeal to authority and/or popularity is illegitimate in scientific discussion.

This is the bait of the bait-and-switch. Interestingly enough, while it is fine as far as it goes in terms of scientific discussion or disagreement between practicing scientists, it breaks down when dealing with discussion between amateurs. That is, even the bait of this bait-and-switch ends up looking shaky.

It is simply a fact that we do not have the time or the inclination to become experts in every single field of science - indeed, even professional scientists can be said to be experts in their speciality alone, and are at best well-informed non-experts with respect to other fields.

As a result, we will often find ourselves in a position where we must choose to defer to some authority or another when attempting to evaluate the merits of a scientific claim, or set of such claims. As a general rule, we rationally defer to those with the most training, practice, experience, and knowledge pertaining to the claim - to the experts, in other words. The higher the stakes (that is, the greater the penalty imposed for choosing poorly), and the greater the expertise required to properly assess a claim, the more rational it is to defer to expert opinion: one consults an oncologist when looking to be treated for cancer, not a plumber, nor a pilot, nor a particle physicist.

If there are multiple competing claims requiring assessment, then, the existence of a scientific consensus is, for the layperson, a useful benchmark for determining the likelihood that a claim is correct. As a simple matter of probability, when the vast majority of scientists (who are notoriously contrarian and competitive - one does not win the Nobel prize for simply confirming what we all knew already, after all) in a specific field agree on the correctness of a claim, they are far more likely to be correct than the remainder. This likelihood increases as the proportion of scientists accepting the consensus position increases, and as the scientists espousing alternate positions become increasingly more fractious (that is, as the number of alternate positions increases even as the number of scientists endorsing them reduce in number or are stable).

(Indeed, owing to the fact of the limits of expertise, most sources I have seen defining the appeal to authority fallacy include the caveat that the fallacy is only committed when one makes an appeal to misleading or inappropriate authority, with well-defined criteria to distinguish between fake experts and the real deal.)

Bottom line: "The race is not always to the swift, nor the battle to the strong, but that's the way to bet."

Premise #2

Arguing that a scientific claim is correct on the basis of the existence of a consensus by subject-matter experts amounts to an appeal to authority and/or popularity.

Now we come to the switch.

As far as I can see, virtually every commonly-accepted fact now taught as routine in science classes amounts to a consensus of relevant experts:
  • There’s a scientific consensus about quantum physics.
  • There’s a scientific consensus about special & general relativity.
  • There’s a scientific consensus about plate tectonics.
  • There’s a scientific consensus about electromagnetism.
  • There’s a scientific consensus about evolution.
  • There's a scientific consensus about the Standard Model of particle physics.
  • There's a scientific consensus about the hydrological cycle.
  • There's a scientific consensus about how objects have colour.
  • There's a scientific consensus that vaccines are a safe and effective medical treatment.
  • There's a scientific consensus about the Big Bang.
  • There's a scientific consensus that the HIV virus causes AIDS.

To paraphrase a common "catch-phrase" used by Charles Darwin: "Many other facts could be given."

The reason these, and more controversial examples, have come to be consensus positions, is not because scientists took a vote. It's because, firstly, these positions are what the evidence has lead them to accept, and secondly, these positions have survived multiple tests and attempts at falsification.

In short, whenever anyone refers to the existence of a scientific consensus about a topic, or defends a claim because it is supported by a scientific consensus, they are using "scientific consensus" as a shorthand for "massive, perhaps even overwhelming, body of theory, experiment, and empirical observation supporting this position or claim, and ongoing failure of attempts to falsify or refute it".

Bottom line: When practicing scientists who, after all, earn renown, rewards, and greater opportunities for research by breaking the mould, overturning the old, and discovering the new and exciting, are forced into near-unanimity about the correctness of a claim or set of claims, you have to be pretty obtuse to go around insisting that such near-unanimity is meaningless, because evidence. After all, it's evidence that has led to the consensus to begin with.

Shorter bottom line: The consensus of experts follows necessarily from the consensus of evidence.

Premise #3

Historically, notable scientists have overturned or broken down existing scientific consensus - and it is this overturning which has made them notable.

To be honest, at this point we can probably pack up with this deconstruction. However, even this premise does not, I dare say, stand up as well to scrutiny as one might think.

As I stated above, scientists have strong personal and professional incentives to overturn older science. However, apart from minor adjustments to mainstream scientific positions (such as, say the determination that H. pylori were key to the formation of duodenal ulcers, contrary to previous conclusions), I can't really think of very many large, well-accepted theories or aggregates of scientific facts that have actually been overthrown, much less by "only one investigator who happens to be right" in the last 100-200 years.

Let's look at a few famous examples to see what I mean.

Albert Einstein
While Einstein might be one of the very few examples of scientists who actually did heavily modify or even overturn a pre-existing scientifically-based paradigm (Newtonian mechanics), it must be noted that Newtonian mechanics were not eliminated entirely from science (one can, after all, get by quite well in life using only classical physics for dealing with cars, bullets, that sort of thing): merely made a specific limiting case of general & special relativity. Furthermore, the history of the theories of relativity shows that much important work had gone on before Einstein published his seminal papers on the subject, laying the groundwork for it, and after, for experimental and empirical confirmation. Indeed, the parallel developments in physics that led to the current scientific consensus on relativity and quantum physics took decades and the work of dozens of key researchers to bring to fruition, and broke a great deal of new ground which could hardly be said to have been the subject of a pre-existing scientific consensus.

Charles Darwin
It is, I suspect, a common misrepresentation of Charles Darwin's work that he is the intellectual originator of the notion of evolution of organisms. Whether my suspicion bears out or not, it is certainly the case that evolution was widely accepted by many naturalists (as quite a few scientists were wont to call themselves at the time), though perhaps not universally so, well before he published On the Origin of Species. While fleshing out the notion some, Darwin's chief innovation was shedding light on the method by which evolution occurred, that is to say, via natural selection through reproductive fitness. And of course, it would take first the addition of Mendel's studies on biological inheritance, and second the discovery and analysis of DNA, to construct a thorough, and thoroughly modern, scientific consensus on the topic.

Galileo Galilei
Galileo is probably the proto-scientist whose record is most abused by pseudoscientists, cranks, and denialists when it comes to these kinds of arguments. Unfortunately for them, Galileo didn't overturn a scientific consensus with his research into heliocentrism - between the treatises of ancient Greek philosophers and the work of Copernicus, heliocentrism was already widely accepted by other natural philosophers. Insofar as any consensus existed about geocentrism, it was a religious, not scientific, consensus. Much of his other work was too novel to be said to be overturning a consensus of any sort, and at any rate, it is probably anachronistic to even be speaking of a modern scientific consensus in the context of 17th-century Europe's scientific scene.

While new discoveries continue to be made, we now have an existing corpus of well-established science that has survived one hundred to one hundred and fifty years of the harshest testing. As such, we can surely safely dismiss the notion that "lone wolf" scientists will suddenly overturn existing consensus positions with brilliant thought experiments, to say nothing of the flotsam and jetsam of quacks, crackpots, cranks, and compromised scientists who often inhabit the scientific fringes (but who also tend to inhabit an all-too-unfortunate front-and-centre position within anti-science circles - Andrew Wakefield springs to mind here).

Bottom line: Revolutionary changes in science take decades and countless man-hours of research work to make happen, and when they do happen they rarely completely overturn a consensus position that is based on good science. Indeed, as the case of Newtonian vs. relativistic physics shows, it seems more likely as time passes that old consensus will (and should) simply be absorbed into the new, rather than being rejected outright.


The argument from consensus is illegitimate.

Since I have (at least in my estimation) shown that the premises underlying the argument against scientific consensus are at best shaky, and at worst entirely incorrect, and, further, I have (at least in my estimation) shown that:

(a) the existence of a scientific consensus is a useful guideline for the scientific layperson, especially if it represents near-unanimity (such as, say, 97-98% of practicing scientists or of papers expressing a view on the consensus position, to take a not-so-random example);
(b) at any rate, a scientific consensus follows from the existing body of theory, experiment, and empirical evidence;

we can say with a very high degree of confidence that the conclusion of the argument against consensus does not follow from its premises, in the first place, and is incorrect on its own merits, in the second. Certainly I for one see no value in it whatsoever. It is, as I stated previously, a bait-and-switch, presenting an irrational and incorrect notion of science (or other fields of empirical knowledge-building, such as history) gussied up with a seemingly-reasonable fa├žade to trick the unwary or uninformed.

1 comment:

  1. Good work.

    Crichton's claim is 'right', but only in the abstract. Of course
    'Science... requires only one investigator who happens to be right, which means that he or she has results that are verifiable by reference to the real world. ... What is relevant is reproducible results.'
    So far as it goes, this is perfectly true. But the 'wrongness' of Crichton's claims far outweigh the 'rightness'; what he is 'right' about is trivial compared to what he is wrong about.

    As you put it, the picture of the lonely and martyred scientific investigator contra mundi is a fiction (it does not even rise, in my view, to the dignity of myth). A scientific consensus arises because, contra Crichton, multiple scientists have tested hypotheses and produced the same reproducible results.

    For, what I omitted of Crichton's argument, postulations such as 'the work of science has nothing whatever to do with consensus,' and 'In science consensus is irrelevant,' and 'There is no such thing as consensus science. If it's consensus, it isn't science. If it's science, it isn't consensus. Period,' are, simply, untrue. The last statement is mere tautology, a circular argument so small you couldn't fit a baby's pinky finger through it.


Idiomatic Composition welcomes your comments and insights.

Commenters may feel free to disagree, even deeply disagree, with one another or with the author on a particular post.

That said, comments are moderated. A minimum degree of civility is expected in any exchange, and comments may be edited or redacted in full (although not deleted) if they are particularly crude, cruel, off topic, or insulting (or spam, it almost goes without saying).