Is Separation of Concerns an Ethical Imperative?

Durkheim, Dijkstra, and the Oracle

Ethics as a study of how one ought to act is fundamentally constrained by the is/ought distinction: it is impossible to make an ethical judgment without knowing what ought to be. Science, on the other hand, concerns itself with what is. The separation between these two fields of human endeavor does not necessarily prevent scientific ethics: one better establish what is, if only to identify precisely what ought to be.

When Émile Durkheim decided to formally establish sociology as an independent discipline, his central claim was that social facts exist, that they can be studied empirically, and that they cannot be reduced to other kinds of facts, for example, to psychological, biological, or economic ones. The last distinction turned out to be very important, otherwise the field of sociology could have been claimed by psychologists or economists! While the validity of the precise terms on which Durkheim defined social facts can be debated, his approach was generally successful: it made Durkheim famous and let him develop a uniquely perceptive view on religion that influenced a number of 20th century thinkers.

In computer science, there is a design principle called separation of concerns (SoC). Edsger W. Dijkstra formulated it a few years after the fateful 1968/69 NATO conferences where a so-called software crisis was openly discussed. Looking back at the decades that followed, the SoC principle was foundational to resolving this software crisis, as well as in establishing software design as an independent engineering discipline. SoC is the core principle behind the wildly successful object-oriented programming paradigm. Nobody writes complex computer software today without carefully thinking about SoC, in one form or another.

Dijkstra, besides being a brilliant computer scientist, was a deep thinker who was well aware of how SoC could relate to science in general. In a very interesting essay, On the Role of Scientific Thought, he describes the SoC principle as being necessary for “effective ordering of one’s thoughts”:

Let me try to explain to you, what to my taste is characteristic for all intelligent thinking. It is, that one is willing to study in depth an aspect of one’s subject matter in isolation for the sake of its own consistency, all the time knowing that one is occupying oneself only with one of the aspects. We know that a program must be correct and we can study it from that viewpoint only; we also know that it should be efficient and we can study its efficiency on another day, so to speak. In another mood we may ask ourselves whether, and if so: why, the program is desirable. But nothing is gained — on the contrary! — by tackling these various aspects simultaneously. It is what I sometimes have called “the separation of concerns”, which, even if not perfectly possible, is yet the only available technique for effective ordering of one’s thoughts, that I know of. This is what I mean by “focussing one’s attention upon some aspect”: it does not mean ignoring the other aspects, it is just doing justice to the fact that from this aspect’s point of view, the other is irrelevant. It is being one- and multiple-track minded simultaneously.

He recognizes that his principle could clash with certain academic tendencies:

It is no slip of the pen that the above quotation refers to the “effective ordering of one’s thoughts”: the efficiency of our thinking processes is what I am talking about. I stress this pragmatic appreciation because I live in a culture in which much confusion has been created by talking about the so-called “academic virtues” (sic!) with moral, ethical, religious and sometimes even political overtones. Such overtones, however, only confuse the issue.

Dijkstra makes a very interesting analogy where he compares scientific disciplines to software modules:

A scientific discipline emerges with the — usually rather slow! — discovery of which aspects can be meaningfully “studied in isolation for the sake of their own consistency”, in other words: with the discovery of useful and helpful concepts.

He then formulates two requirements of a viable scientific discipline:

…for the separation to be meaningful, we have also an internal and an external requirement. The internal requirement is one of coherence: the knowledge must support the abilities and the abilities must enable us to improve the knowledge. The external requirement is one of what I usually call “a thin interface”; the more self-supporting such an intellectual subuniverse, the less detailed the knowledge that its practitioners need about other areas of human endeavour, the greater its viability. In the terminology of the computing scientist I should perhaps call our scientific disciplines “the natural intellectual modules of our culture”.

All of this brings us to the still ongoing discussion about how ethics relates to the pursuit of truth. A common point made by some scholars is precisely the one against which Dijkstra railed: that academic virtue is inextricably linked with moral, ethical, religious, and political concerns (or ought it be linked?). In a recent Twitter discussion, Australian philosopher Patrick Stokes makes an interesting claim that “ethics has no outside”:

My disagreement with his claim is in the following: this “lack of outside” is irrelevant insofar as the is/ought distinction is unclear. Ethics may not have an “outside,” but it is constrained by the above distinction. What ought to be is the fundamental currency of ethics, and without it ethics becomes impossible. In ignoring this distinction, Stokes implies that science cannot make claims about ought, whether because of some epistemological problem, or simply because Stokes does not want it to. This is a rather radical position masquerading in an innocuous sentence. For Stokes, scientific ethics is either impossible or undesirable. Nothing that science produces can impact the ought, in other words. One would get a similar situation under a theocracy, where all claims to ethical authority implied by ought are located outside of science.

The opposing view is that ethical concerns can “contaminate” pursuit for truth — to which Stokes responds by making an even stronger claim that not only scholarship, but truth itself is inseparable from ethics:

If truth is inseparable from ethics, and if, as I claim above, ethical authority according to Stokes lies outside of science, then this authority also becomes the arbiter of truth while being outside of science. Just what is happening here? Who is the arbiter? The oracle of Delphi?

In Plato’s The Republic, Socrates imagines a society where ethics implies a very different relationship to truth from the one most of us are familiar with. Socrates’ imaginary society operates according to a noble lie perpetuated by the rulers: “for an oracle says that the State will come to an end if governed by a man of brass or iron.” The ethics under such a society (as determined by “men of gold and silver” though deceptively attributed to the oracle) would be in conflict with the empirical truth. Clearly, Plato had a good grasp of SoC.

I find the contrast between Durkheim’s definition of social fact as something relatively independent of other kinds of phenomena and the currently popular conflation of ethical and factual to be fascinating. It implies a radical break from the traditional positivist view of science, something that slipped by us in the second half of the 20th century without many people noticing. To clarify, I believe that Durkheim was entirely correct in wanting to separate social factors from hereditary ones. This does not imply that the empirical social researchers whose work consists of building mathematical models to explain social phenomena are free to remove biological factors from consideration, however.

It is impossible not to make one final note. If one agrees with Durkheim and Dijkstra (and in the absence of a strong reason not to, one probably should, as these were very exceptionally wise individuals) about what science ought to be, then a certain separation between truth and ethics becomes itself an ethical imperative.