$1.1 An overview. Many years ago, during the course of a third-year law school seminar in the taxation of corporation reorganizations, our professor commended to us what he called the most amusing opinion he had ever read in a field, as he put it, "that does not abound in humor." The opinion turned out to be a hopelessly confused judicial decision in which an eminent jurist demonstrated his total incomprehension of the purpose and operation of the tax treatment of certain corporate transactions.

In retrospect, I no longer find it diverting that a scholarly judge of high intelligence, writing after careful study and much concentration, cannot make sense out of the statutory expression of the rules governing certain conduct. The growing impenetrability of much federal legislation is not wholly, or even primarily, caused by its draftsmen. It mirrors the increasingly complicated ways in which government intervenes in private and public activity that itself continues to gain in sophistication.

In the face of this ever-burgeoning complexity, the draftsman has a very special responsibility. It is the legal analog of the Hippocratic injunction on the practice of medicine: First, do no harm. It may be stated as follows: Let's not make things more complicated than they have to be. This is not as novel an idea as it may appear. In 1817, Thomas Jefferson wrote to a Mr. Cabell:

I should apologise perhaps for the style of this bill. I dislike the verbose & intricate style of the modern English statutes, and in our revised code I endeavored to restore it to the simple one of the antient statutes, in such original bills as I drew in that work. I suppose the reformation has not been acceptable, as it has been little followed. You however can easily correct this bill to the taste of my brother lawyers by making every other word a 'said' or 'aforesaid,' and saying every thing over 2. or 3. times, so as that nobody but we of the craft can untwist the diction, and find out what it means; and that too not so plainly but that we may conscientiously divide, one half on each side.

Since Jefferson's time, the draftsman's passion for the turgid and redundant has somewhat abated. Today, it is not the abundance of whereas's and aforesaid's that interfere with the intelligibility of so many of our statutes. Instead, it is the poorly organized, convoluted, or otherwise slovenly treatment of statutory concepts that demand precision.

This text will not solve that problem; it cannot by itself turn a law student or a lawyer into a legislative draftsman. If you are willing to devote the time needed to read it with care and do the assigned exercises, if possible under the guidance of an experienced draftsman, you will nevertheless be on the way to solving the problem for yourself in your own work.

We will study legislative drafting from the perspective of a draftsman for a federal agency. Virtually all major programs of federal financial assistance, and most of the significant regulatory statutes, have in their ancestries a proposal made to Congress by an executive agency, customarily in the form of a draft bill. Generally speaking, these proposals are developed with greater formality than bills written within Congress. In the Department of Health and Human Services, which is probably unrivaled within the Federal Government in the extent and diversity of its legislation, a legislative proposal is commonly reduced to written specifications. This arrangement seems optimal. It compels the responsible policy officials to refine the content of a proposal, at least in a preliminary fashion, before they may look to the draftsman for preparation of a draft bill; yet the process usually offers the draftsman an opportunity to participate in the proposal's formulation.

The last of your tasks, as a legislative draftsman asssigned by a federal agency to prepare a major draft bill for submission to Congress, is actually writing the bill. If you have done the necessary preliminary work, it is the task that is the least timeconsuming. Here are the steps you should take, in chronological order, before you begin to write:

(1) General policy review. You should review drafts of the "decision memorandum" that explains the proposal in general terms to the agency head and asks his approval for its further development.

(2) Issue refinement. After the agency head endorses the initial general proposal, you should help to refine, develop, and present issues for his subsequent decision.

(3) Preparation of legislative specifications. You should advise on how to prepare the "legislative specifications", i.e., the written expression of the detailed policy decisions that you must incorporate in the bill.

(4) Clarification of legislative specifications. Upon receiving the specifications, you should clarify them

through telephone calls, meetings, or memoranda.

(5) Preparation of drafting outline, if necessary. If the proposal involves amendments to current law, you should prepare a drafting outline for your own use that specifies each section of current law that must be amended, and describes how it is to be amended. If the proposal is for a wholly new statute, you should outline the contents of each section of the draft bill.

If you have taken these steps, you are ready to write the first draft of the bill, circulate it within the agency for review, revise it to take account of comments, circulate a second draft, and so on until a draft is finally agreed to.

§1.2. Guides for evaluating the adequacy of legislative specifications. When you receive a set of legislative specifications approved by an agency policy official, you know that the awful moment has arrived when you are actually called upon to do something. This moment is especially terrifying when, as often happens, the time reserved for the draftsman has been eroded by delays in reaching decisions on the specifications. Often the policy debate would not end at all except that over the horizon there comes into view some event a presidential message, a congressional committee hearing, a subcommittee bill mark-up-that irresistibly compels argument to yield to action. All eyes then turn upon you and you are told that unless you prepare your bill within X days (X days always being fewer days than the job demands) the government will fall to its knees.

At Appendix A you will find specifications for a bill to deal with "domestic violence". These specifications are reproduced without change from the files of the Legislation Division of the Department of Health and Human Services (formerly the Department of Health, Education, and Welfare). Please read them carefully.

Notice that the specifications at no point reveal what "domestic violence" is. The uninitiated could readily conclude that the specifications describe a proposal for services to those injured in urban street fighting or a program of law enforcement assistance to deter civil riots. This omission illustrates an important truth. Legislative specifications are usually vague and incomplete. In part this is attributable to the policy officials' reliance on the draftsman's knowledge and judgment; even more so, this reflects the embryonic stage of the policy officials' thoughts. The draftsman must now engage the policy officials in a colloquy to inform himself exactly as to what he must do. Often this colloquy will reveal to the policy officials that they are not sure, beyond general objectives, what they themselves want to accomplish.

For example, although you might infer that the specifications look toward a bill intended to reduce wife-beating you do not know whether the bill is to make federal funds available for projects dealing with child abuse or parent abuse. In all probability you will find that on these major issues the policy officials have a definite view, which they merely neglected to express in the specifications. If you were to ask, however, whether they wish a federally supported domestic violence project to assist an adult male who has been injured by his adult brother with whom he shares an apartment, the answer may be less certain.

In extreme cases, specifications are so incomplete as to be unacceptable. Most of the time, though, they are sufficient to enable an experienced draftsman, after clarifying a handful of issues, to write a first draft that contains at least something on every item with which they deal. Necessarily, this means that the draftsman will have to anticipate ultimate policy decisions on innumerable small matters.

On the question of when to draft, I follow these guides:

(1) Essential concepts. If the specifications are obscure on an aspect of the proposal that would be time-consuming or otherwise hard to draft-something that might take an experienced draftsman more than an hour, let us say I prefer to obtain guidance on what precisely is intended.

(2) Boilerplate. If the specifications are obscure on "boilerplate", that is, those portions of agency programs that tend to show up in similar form from statute to statute, I do not seek guidance. Instead, I draft what I think will be an appropriate set of provisions.

(3) Other. If the specifications are obscure on matters that do not fall readily under one of the two preceding rules, what I do depends upon the time I have available to prepare the draft and my feeling for the material.

Some draftsmen prefer to draft immediately upon receiving specifications, simply guessing at obscurities. They argue that the test of whether a draftsman understands an idea is whether he can write a provision expressing it. During the writing itself he will discover gaps and ambiguities even in specifications that at first seemed complete and clear. If the technique of early drafting wastes some time and energy, these may be more than repaid, they contend, by the insights that early drafting will give the draftsman into the demands of the job: insights that he will need in order to make his initial meeting with the policy officials on the proposal as productive as possible.

Of course, if you receive specifications about whose subject matter you have no prior knowledge, nothing will be clear enough to draft. To guard

against this, the first two steps that §1.1 lists as preceding the actual bill writing are general policy review and issue refinement. Both involve use of the draftsman before the specifications are written. Contrary to the views of those (usually lawyers) who believe that the legal discipline especially fits one to formulate social policy, the early involvement of the draftsman in general policy review and specification refinement is less in tribute to his potential contribution to those enterprises than to a need to give him early exposure to policy thinking. As the draftsman, you need this exposure to understand the issues and how your clients approach them. It will enable you to guide yourself by both your awareness of what the policy officials are seeking and your recollection of the choices that they have rejected. Together with your program knowledge of what is administratively feasible, this early involvement will enable you to perform creditably at high speed when the specifications arrive.

$1.3 Applying the guides. Using the domestic violence specifications as an example, you probably should not attempt to draft a bill until you obtain a decision on whether all authority, consistent with what was the usual HEW practice, is to be vested in the Secretary rather than in subordinate officials or entities. Before beginning to draft, you will also want guidance on the citizen participation requirements, because the specifications may contemplate an elaborate system of public hearings leading toward the preparation of an annual state services plan and participation of volunteers and nonprofit private groups to implement it. (See, for example, section 2004 of the Social Security Act.) On the other hand, you should not wait for answers to questions on the need for sanctions should a state fail to comply with its plan assurances. You should simply write in a provision such as this:

If the Secretary, after reasonable notice and an opportunity for a hearing to the State, finds that the State has failed to comply with any requirement under this Act, he shall notify the State that further payments will not be made to the State under this Act until he is satisfied that there will no longer be any such failure to comply, and until he is so satisfied he shall make no further payments to the State.

Your explanatory memorandum circulating the first draft for review should alert the addressees to your treatment of the sanctions issue. They will tell you if they have different ideas.

You may well ask why you should waste any time at all in drafting on the basis of your best guess as to the intentions of the specifications. Admittedly, it might be more efficient to remove all uncertainties

before drafting, but this is generally not practical. Specifications are typically fluid and different policy officials often have different ideas on details. A draft bill is a marvelous instrument for concentrating the mind of the policy maker; it usually precipitates many specifications changes that might not be thought of otherwise. The draftsman observes that today's decisions, hastily made, are tomorrow's decisions hastily reversed. Time spent on perfecting a first draft is thus usually time wasted. Moreover, even when major policy changes are not in prospect, policy officials remain an impatient breed, especially when awaiting the work of others. They ordinarily prefer to review an imperfect draft bill rather than countenance the delays that are sure to attend the draftsman's effort to resolve all policy issues before drafting. As far as technical matters go, you can polish the draft while the policy makers are studying it. There is no need to delay their review because you want to perfect the bill. As in other areas of the law, the most precious commodity in drafting is time.

§1.4 Determining where to clarify specifications. When reading specifications, you must constantly ask yourself, "If this were a statute addressed to me, how exactly would I go about carrying it out?" Put yourself in the place of one who must administer the specifications once they become law. Think through in detail the specific actions that you would have to take. You need not translate all of these specific actions into bill language; it is not your objective to write as detailed a bill as you are able to imagine. Your purpose is simply to assure yourself that the specifications will result in a statute that is unambiguous and capable of being followed.

If you look at specifications this way, you will be able to write a bill that facilitates the actions of those who must implement it. To take a trivial example of the neglect of this principle, the morning Washington commuter who drives west along Independence Avenue daily encounters illuminated signs apparently instructing him to "Use All Lanes."

$1.5 Examples of unclear specifications. Let us apply this way of looking at specifications to the domestic violence specifications at Appendix A. To begin with, notice that they announce that state grants are to be "distributed by formula based on population with a floor of $100,000 per State." Assume that this phrase becomes law and that Congress appropriates $20 million for 1980. Some administrator must now figure out who gets what. The first step is easy: the specifications (which we now deem to be the law) allot 65 percent of the appropriation for the state grant program. Sixty-five percent of $20 million is $13 million.