I do not write codes of ethics. I help companies write them. I have helped several dozen companies as a paid consultant, more in less formal ways. Every company was different. Some came in with only the general idea that they should have a code; some came in with a rough draft they wanted me to criticize; some came in with a draft already printed and bound.

No matter at what stage of drafting they were, I would begin with the same advice: I should not be the one to write their code of ethics; they should. They know their culture, the ethically sensitive areas in their operation. Next, I would tell them that writing a code of ethics is not just puffing words on paper. Writing a code is at least as much process as outcome. They need to think carefully about how they will get "buy in" from every part of the company, how the code will guide practice, and how the code will be used to teach good practice.

What then? The Bentley Center for Business Ethics keeps a collection of corporate codes. If a company thinks it needs a "benchmark", I send them a few of these, telling them that they will have to adapt them to their own environment. Later, as the draft develops, I make comments, for example, "Too much legalese, write so that people in your company can understand it." I look at the content, readability, length, and omissions ("Why nothing about conflict of interest?"). I may suggest including pictures, bullets, big type, a question-and answer section off to one side-things to make the code "user friendly".

A code should be more than the rules of the road; it should include a statement of the company's core values. If the company does not have such a statement, then the first words they put on paper should be a statement of core values. The code should be organized around the core values.

Among the things that make a code "user friendly", the most commonly overlooked is advice on how to use it. The code should include a telephone number for getting interpretations of the code, a procedure for raising an ethical issue ("first go to your supervisor, then to..."), and even a procedure for suggesting changes in the code. The code should also include an ethical decision-making model ("Step 1: Check your facts, Step 2....").

There are also subtler issues of tone to consider. For example, I was once asked to look over a code titled "Our Responsibilities". Good title, but every rule that follow,i J began with "it is your responsibility to..." I suggested changing every "your" to "our", understanding that the rules would apply to everyone in the company, top management included.

Helping with the process of writing the code is as important as helping with the code's form and content. One thing I tell every company that comes to me is that a code needs someone within the company who writes well - a Thomas Jefferson - to give it order, clarity, and precision. But the code should pass through many stages before that local Thomas Jefferson gives it a "final" shape. The code should not be written by one person, or even just one group (such as senior management or the public relations department), and then shipped out. Any task force drafting a company's code of ethics should involve representatives from every level of employee.

"Every level" means every level. I once had a company come to me with a plan for drafting a code that they thought met this standard. As I looked through the plan, I realized that it omitted one group, unionized employees. I asked why. The omission was not an oversight. They (the company's officers) had thought about including union representatives but decided not to because they thought the union would not be interested or would be hostile to the project. I told them they should at least try to bring the union in. After all, it represented a majority of the company's employees. The company eventually agreed to try. The union surprised them; they participated willingly and well. Later, the company told me that including the union was the best advice I gave them.

Once the local Thomas Jefferson has polished the draft, the company generally wants to print it on slick bond, with a stitched binding and handsome cover. I tell them to print it on cheap eight-and half-by-eleven paper, stapled inthe upper left cornor, with "DRAFT" in large type at the top of every page. However finished the code seems to those who wrote it, the company as a whole should first see it as a draft sent out for comment.

That the code is sent out only in rough draft does not mean that it should not be used. It should be used right away, for example, in the company's ethics training program. Using "pilot" versions of the code is the best way to find out how clear it is, how relevant to actual problems, and how appropriate are its recommendations. The training program should give employees ample opportunity to work with the code and make suggestions for its improvement. Like any good ethics training program, one testing the code should be organized around " cases" that might arise at the company.

The code should go through several pilots before it goes out in its final form. "Draft" should be noted at the top of the code each time until it is finalized. Employees should feel they have had a part in drafting the code; they cannot feel that unless they have had a part in drafting it. They must see the code changing in response to their input.

If employees are not brought into the process in some such way as I have suggested, they will be turned off. The code will seem something "they" have imposed on "us". That's not what ethics should be. Ethics should be part of an organizational community. Everything should be done to make employees see that having a code of ethics can strengthen the ethical environment in which they work, as well as protect the company legally. Everything should be done to make employees understand that the code is subject to change, revision, and renewal - and that they will have a part. So, in a sense, the code is never finished.

Not every attempt to write a code of ethics ends in success. One example of "failure" may suffice to make clear what can go wrong. Once I was asked to consult at a company whose leaders seemed committed to an ethics program. After I gave my usual advice, they began the process of creating an appropriate task force. Letters were sent out to employees indicating top management support. Then, somewhere along the line, someone important must have asked, "Do we really need a code of ethics? Why not just have a compliance program?"

Whatever happened, happened behind closed doors. So, I don't know what the reasoning was. Perhaps someone realized that the ethics program would take resources needed for another department. Perhaps the board of directors realized that an ethics program would change the company in ways they did not want it changed. But, at some point, the company decided they didn't want an ethics program.

All this was very disheartening for me. I had already spent lots of time on the program. I had talked with employees and persuaded them that an ethics program was a good idea. The employees were excited. Commitments to the importance and benefits of the program had repeatedly been made from the highest levels of the company. Then, someone come along and stuck a needle in the balloon.

But, most disheartening, I thought the company was making a mistake. A company without an ethics program is a company at risk. How much risk? I recall another company that began its ethics program by doing a survey of employees to gauge the company's ethical climate. Top management found the results shocking. They had not appreciated how little ethical guidance they had been giving - or what effect the absence of guidance was having.

I had no formal preparation for this sort of consulting. I began my career at Bentley more than twenty-five years ago with a Ph.D in philosophy. As chair of the Philosophy Department, I tried to "hook up" philosophy with business ethics. After a while, I sounded knowledgeable enough that business professors began to ask my help in their consultations. I have been fortunate to have good partners who helped me gain a great deal of experience before I tried consulting on my own. The opinions expressed here are the fruit of that experience.