Several times a year CSEP receives an inquiry from another professional society: "Do you have anything on how to write a code of ethics?" Each call reminds us of how little is in print on that subject. Those who write professional codes rarely publish much about the process. Why? Doubt that others can learn from them? Too busy to do what academics routinely do - observe, record, analyze, and publish? Too worried about confidentiality?

Preparing this issue of Perspectives left such questions unanswered. I listed everyone I knew to have helped write a code of ethics, and then asked one after another to write something helpful about code writing. A few declined because of other commitments, but the refusal rate was not high, nor was there any obvious pattern to the refusals.

First Professional Code?
I also asked Robert Baker, who has not written a code of ethics, to write about a favorite subject, the history of such codes. His piece, though no help understanding why so little is written, gives historical context to the three pieces following it.

Baker dates the first code of professional ethics between 1794 and 1808, explains why professional codes replaced the older ethics of oath and gentlemanly honor, and even tells us something about the origin of what he thinks was the first professional code.

A hospital asked Thomas Percival, a physician there, to draw up rules to guide its physicians and surgeons. Having done as asked, Percival generalized the guide, creating something like a textbook on the law and etiquette of practicing medicine in England (Medical Jurisprudence, 1794). After eight years of hearing what readers thought of that work, Percival revised it, publishing the result under the new title Medical Ethics, and promptly died, leaving behind no equivalent work on how to write a code of ethics. Indeed, I am less sure than Baker that Percival ever wrote a code of ethics. Why?

Percival's Achievement
The word "code" comes from Latin. Originally, it referred to any wooden board, then to boards covered with wax used to write, then to any book, and finally to an authoritative systernizing of lawsbecause the Emperor Justinian called his book- digest of Roman law a "code". Justinian's code (529 AD) differed from an ordinary digest or other compilation of law in one important way: his was enacted into law, replacing everything that came before. The Justinian Code began Roman law anew.

Since then, anything sufficiently like Justinian's large-scale rulemaking has also been a code. For example, the spy's "secret code" (cipher) is a code because it is an authoritative system of written rules (something analogous to laws), though only concerned with converting one set of symbols into another (ciphering and deciphering). "Computer code" is code because it resembles the seeming nonsense spies wrote.

The subtitle of Medical Ethics (which Baker omits) begins: "A Code of Institutes". This odd expression suggests that (even in 1803) Percival did not think of himself as publishing a code (strictly speaking). The term "institutes", like " code", goes back to Justinian. Justinian's Institutes was a textbook in "jurisprudence", a digest of laws and legal opinions designed for law students. Percival's odd combination of "institutes" with "code" permits only an extended sense of " code", for example, code as mere systematic treatment of its subject.

Because Percival lacked the authority to enact rules for his profession, Medical Ethics must be a digest or textbook, not a code (strictly speaking). Only the rules Percival wrote for his hospital were (once adopted) a code.

Could Percival have written a code of ethics by putting an unwritten code into writing? That is a question of how far to stretch the analogy with Justinian's code. An analogy stretched too far will snap, harming where it should help. If &&unwritten" simply means having an authoritative form not yet in writing, then codes can be unwritten. The constitution of medieval Iceland was unwritten in this way. The Icelanders, being illiterate, preserved its exact text in verse. Each year, at the opening of the national legislature, a bard sang the verses, just as he would a saga. Any dispute about the wording of the constitution could be settled by singing the appropriate verse.

The medical ethics of Percival's time does not seem to have been unwritten in this way but in another: before Percival wrote his book, medical ethics had no authoritative formulation. Since the point of codification is to give law (and, by analogy, any similar system of guidance) an authoritative formulation, a code without an authoritative formulation would seem to be no code at all. Any code, including any code of professional ethics, must have a set form (written or oral).

Ethics?
Did Percival at least write rules of professional ethics? That depends on what "ethics" means here. The term "ethics" has at least three senses in English. It can be a) a mere synonym for ordinary morality, b) the name of a field of philosophy (the attempt to understand morality as a rational undertaking), or c) the name for some special, morally permissible standard guiding members of a group (for example, the ethics of gentlemen).

It is only in this last sense that professions have their own ethics. Legal ethics, for example, governs lawyers (and no one else). A profession's code states in authoritative form the special standard governing its members, a morally permissible standard everyone in the profession wants everyone else in the profession to follow even if that would mean having to follow it too. A professional standard is morally binding because it constitutes a practice from which each participant benefits if others do their share, a practice each voluntarily enters by claiming membership in the profession (and having that membership acknowledged by other members). Professional ethics is voluntary in a way neither law nor the regulations Percival wrote for the physicians and surgeons of his hospital are. Was that voluntariness what Percival meant to convey when he substituted "medical ethics" for "medical jurisprudence"?

What follows from these definitions is that, while Percival may have invented the term "medical ethics" and certainly wrote "institutes" of ethics, he could not write a code of professional ethics. Writing such a code had to wait until a sufficiently authoritative body, the profession itself, adopted something like Percival's standards. By Baker's reckoning, that could not be before 1808, when the Boston Medical Society adopted a code of ethics for its members.

How should one write a code of professional ethics today? While there may be no single answer to that question, we may learn much from comparing what professional societies have done recently.

Professions and Business
Donald Gotterbarn describes the writing of two professional codes, the "Code of Ethics and Professional Conduct of the Association of Computing Machinery" (1992) and the "Software Engineer's Code of Ethics and Professional Practice" (1999). Writing the ACM code seems to have been a relatively informal process. Writing the SE code, the work of two large societies, was much more formal, involved many more people, and took much longer.

Despite these (and other) important differences between the two codes, there are several striking similarities. The actual drafting of each was primarily the work of a few people. The editing and revision had somewhat wider participation, but only ratification involved the general membership or even their elected representatives. Gotterbarn prefers the more structured process by which the SE code was written, especially the formal reviews that brought in people who might otherwise not have come in until ratification.

Our next piece on writing a professional code may embody in full the ideal Gotterbarn only sketches. In "Developing a Code of Ethics for Early Childhood Education", Kenneth Kipnis and Stephanie Feeney describe a process that, almost from the beginning, involved a substantial part of the profession. For example, over 800 pre-school teachers sent in answers to the first ethics questionnaire. Later, each draft of the code was not only published for comment but tested by hundreds of 11 end-users"- -much as "beta" software is. (is there more to learn from the analogy between codifying ethics and writing software?)

In our last piece, Michael Hoffman summarizes two decades of helping businesses write codes of ethics. Business codes may seem too for from professional codes to deserve mention here - yet Hoffman's conclusions generally reinforce those of Gotterbarn and Kipnis-Feeney. For example, Hoffman stresses that, while every code needs one "Thomas Jefferson" to give it coherence, those expected to follow the code should be brought into the process as early and as often as possible.

Somewhat more is in print on how to write a code of business ethics than on how to write a code of professional ethics. While professional societies would certainly benefit from a systematic comparison of professional codes like that the Ethics Resource Center has done for business, they might also benefit from what is already in print about codes of business ethics. Code writing is one activity business and professional ethics have in common.

- MD