Consider this dialogue:

A: “All fish breathe through gills rather than lungs.”

B: “But whales are fish, and they breathe through their lungs.”

A: “Whales may look and seem like fish, but they aren’t truly fish because they breathe through their lungs.”

To anyone with a rudimentary knowledge of biology, A’s reasoning here must seem sound. Yet among some philosophers with a scientific bent, the structure of the reasoning A employs is often criticized as a logical fallacy. It is specifically known as the “No True Scotsman” fallacy, popularized by Antony Flew in this passage from his Thinking About Thinking:

Imagine Hamish McDonald, a Scotsman, sitting down with his Glasgow Morning Herald and seeing an article about how the “Brighton Sex Maniac Strikes Again”. Hamish is shocked and declares that “No Scotsman would do such a thing”. The next day he sits down to read his Glasgow Morning Herald again; and, this time, finds an article about an Aberdeen man whose brutal actions make the Brighton sex maniac seem almost gentlemanly. This fact shows that Hamish was wrong in his opinion but is he going to admit this? Not likely. This time he says, “No true Scotsman would do such a thing”.

What Hamish McDonald does in this hypothetical case sounds like a fallacy. It might indeed be one – but that depends on another fact about the case which is not mentioned in Flew’s paragraph. For as the case is stated, Hamish’s pattern of reasoning is exactly the same as the scientist above, making a perfectly valid inference about fish. If the structure of Hamish’s claim requires us to view it as a No True Scotsman fallacy, then the scientist’s claim must also be viewed as a No True Fish fallacy.

The important fact in question is this: does Hamish (or A!) in fact keep changing his definition in order to evade? Or did he already have a normative definition in mind of what a (true) Scotsman is, which simply happened to be different from the definition in common use? In Flew’s example as stated, it could very well have been the case that Hamish already believed a Scotsman was a man who was not merely Scottish by birth or citizenship but conducted himself according to a particularly Scottish standard of propriety, one which included not conducting oneself with brutality. In such a case, “this fact” of the brute being from Aberdeen does not in fact disprove the claim that no Scotsman would do such a thing.

One can of course dispute such a definition, but there is no fallacy about it whatever – any more than there is a fallacy in the definition of fish as not including whales. Look at a whale – it lives its life in the water and it is shaped like a fish. Common sense, in the sense of untrained inferences from perception, indicates that a whale must be a fish. As is so often the case, common sense is wrong. And common sense tells us that anyone who has spent his whole life in Aberdeen must be a Scotsman. But if Hamish does have an alternate understanding of Scottishness, of the sort discussed above, then we would do well to listen to him and at least think about that alternate definition, rather than laughing at him for committing a fallacy when he is doing nothing of the sort. (One may note in passing that there is in fact a widespread usage of the term “true Scotsman” which has little to do with Scottish ethnic or geographic background.)

The reason all this matters is that the definition of “a true X” employed in such cases is often a normative, value-laden definition: a true Christian would not be miserly toward the poor, a true American would support freedom of speech even for those forms of speech he despises. The so-called “RationalWiki” has a page on the so-called fallacy which currently proclaims a typically egregious example: “Religious apologists will repeatedly try to use the No True Scotsman argument to distance themselves from more extreme or fundamentalist groups, but this does not prevent such extremists actually being religious – they themselves would certainly argue otherwise.” The assumption in here, unfortunately shared by far too many in religious studies, is that someone calling himself “religious” makes him religious in general, and someone calling herself a Christian or a Buddhist makes her a Christian or a Buddhist. Which works fine for people who think those traditions are complete bullshit and are simply looking for ways to classify all those idiots out there, but is not fine for those of us who think there might actually be some truth in the traditions and are hoping to find it.

Those “extremists” here being cited as an authority would certainly say that they are true Muslims (or true Christians) – but they certainly wouldn’t agree that everyone claiming to be a true Muslim is indeed one. What counts as a true Muslim or a true Christian is itself part of what’s being debated.

And that is exactly as it should be. Those of us seeking knowledge of value, of good and bad, need to respect the connection between “an X” and “a good or true X”. This connection is often what bridges the presumed gap between “is” and “ought”, as in Alasdair MacIntyre’s examples in After Virtue:

From the premise ‘He is a sea captain’; the conclusion may be validly inferred that ‘He ought to do whatever a sea-captain ought to do’. This counter-example not only shows that there is no general principle of the type alleged; but it itself shows what is at least a grammatical truth—an ‘is’ premise can on occasion entail an ‘ought’ conclusion. From such factual premises as ‘This watch is grossly inaccurate and irregular in time-keeping’ and ‘This watch is too heavy to carry about comfortably’, the evaluative conclusion validly follows that ‘This is a bad watch’. From such factual premises as ‘He gets a better yield for this crop per acre than any farmer in the district’, ‘He has the most effective programme of soil renewal yet known’ and ‘His dairy herd wins all the first prizes at the agricultural shows’, the evaluative conclusion validly follows that ‘He is a good farmer’. Both of these arguments are valid because of the special character of the concepts of a watch and of a farmer. Such concepts are functional concepts; that is to say, we define both ‘watch’ and ‘farmer’ in terms of purpose of function which a watch or a farmer are characteristically expected to serve. It follows that the concept of a watch cannot be defined independently of the concept of a good watch nor the concept of a farmer independently of that of a good farmer; and that the criterion of something’s being a watch and the criterion of something’s being a good watch.

This connection between an X and a good X is no mere Western conceit; it shows up all over the place. It is at the heart of Confucius’s “rectification of names” (zheng ming 正名): there is a proper social order when fathers father and kings “king”, when they act in the manner that their names indicate they should act. Nor is this account of the connection between an X and a good X necessarily conservative, fixing existing usage. One can use it just as well as a critique: you say you’re Muslims, but you’re not true Muslims if you promote compulsion in religion. Chapter 26 of the Pali Dhammapāda uses this approach to critique the brahmins of the age, who think that their birth into the brahmin caste makes them superior people. The chapter retains the value-laden concept of “brahmin” as a superior person, but detaches that value-laden concept from the usual descriptive concept of birth. In the Eknath Easwaran translation (easily available online):

Not matted hair nor birth
Makes a man a brahmin,
But the truth and love for all life
With which his heart is full.

Of what use is matted hair?
Of what use a skin of deer
On which to sit in meditation,
If your mind is seething with lust?

Saffron robe, outward show,
Does not make a brahmin,
But training of the mind and senses
Through practice of meditation.

Not riches nor high caste
Makes a man a brahmin.
Free yourself from selfish desires
And you will become a brahmin.

There is no fallacy in any of this. If we declare this normative element in words a fallacy, we remove one of our most important grounds for making claims about goodness and badness, and moves us one step closer to the nihilism of the early 20th-century analytic philosophers, who could find no rational basis for making claims about good and bad at all – on the grounds of fallacious reasoning of their own.

EDIT: The original version of this post missed some quotation marks in the hyperlinks, thus leaving out significant parts of it, and had one sentence – the one beginning “In such a case” – that was phrased wrong, indicating the opposite of what I meant. Fixed. Reminding myself to proofread my posts better…