I

The subject of Prof. Reinsch's paper this afternoon emphasizes another need for accurate drafting. If the initiative is to be made a successful method of legislating, means must be provided for the scientific preparation of initiated measures. Bills must be reasonably within the comprehension of the people if they are to be enacted or rejected intelligently. Errors and "jokers" are less likely to be detected by the whole mass of the people than by committees of the legislature, and, if detected, are more dangerous because the bills cannot be amended in the course of discussion and before final action as they could be in the legislature.

My subject is not the need for or the desirability of better drafted statutes, but the means by which they may be had. The scientific preparation of a statute involves :

1. Knowledge of conditions proposed to be regulated, and determination of the exact evils requiring regulation.

2. Determination of the nature of the regulation required and the precise principles or rules which will effect such regulation.

3. Phraseology of the new principles or rules and of necessary administrative provisions in apt and precise language which will fit them into existing principles of constitutional and statute law and make them reasonably clear to the executive and judicial officers who are to enforce them.

So-called practical legislators are fond of dividing these problems into: (1) matters of substance, which are for the legislator, not for the drafter, and (2) matters of form, which may be delegated to the drafter. The distinction, however, is of little value, for changes in phraseology frequently result in changes in policy. Policies determined upon in conference are often hard to recognize when they come from the pen of the drafter. No such division of the problems of preparing legislation is possible. So-called matters of substance and matters of form go hand in hand, and if the problems of legislation are to be solved wisely and effectively, the legislator and the expert drafter must work together.

1 Initiative and Referendum," by Prof. Paul S. Reinsch, University of Wisconsin. See p. 203, infra.

Prof. John W. Patton, of the University of Pennsylvania Law School, says:

Legislative action, however, should be based upon demonstrated need, careful study of the proposed remedy in substance, of its constitutionality, of the meaning of every word used in a proposed act, with a careful examination of existing decisions as well as statutes. Knowledge of law as well as of the English language is required, and the pen of one who thinks he has a facility for legislative expression should indeed "make haste slowly."1

In workmen's compensation legislation, for example, the legislator, if he performs his legislative duty seriously, must first study existing employers' liability law, and the evils, if any, produced by its operation. He must analyze these evils and consider the possible methods of remedying them, and for this purpose he ought to know and appreciate the methods by which in other states or countries similar evils have been remedied. Having decided that the compensation system offers the best means of doing justice, there remain for his decision important questions of policy involved in working out the details of such a scheme. For example, shall the scheme apply in all employments, in all with certain exceptions, or in certain specified employments selected because of their extra hazard or otherwise? Are all injuries in the course of employment to be compensated, or are certain injuries, such as those caused by an employe's own deliberate act, to be excepted? Upon what basis shall the compensation be computed, and how shall the computation be made, and under what conditions shall it be paid? What shall be the procedure to determine controverted questions? What, if any, administrative organization is required for the proper enforcement of the scheme? Every one of these problems involves the determination of a multitude of detailed questions of policy before the precise limits of the rights and liabilities created by the act are defined in such manner that employer, employe, administrative officer and the court may know when and to what extent the legislature intended that A, an employer, should

1" Festina Lente," Penna. Law Rev., vol. 59, p. 214.

compensate B, his employe, in case the latter is injured in the course of his employment.

The foregoing are frequently described as questions of policy with which the drafter should have nothing to do; they are solely for the legislator. Theoretically, this is true. If all these questions were carefully weighed and decided by the legislator there would be nothing left for the drafter but to put the legislative decision into language. Practically, however, the great majority of these questions of policy do not occur to the legislator until the drafter in the detailed statement of the legislative intent uncovers the numerous instances to which the legislative intent has not been applied.

Determination of these questions of policy by no means completes the legislative task. There remain questions of constitutionality and the selection of devices, such as the so-called elective scheme, to avoid constitutional restrictions; the adjustment of the statutory scheme decided upon to the existing statute law on the same or similar subjects; and, finally, the selection of the language which will carry the statutory scheme into a statute at once constitutional and effective for the purposes for which it was intended.

Mere phraseology of a statute is itself a difficult task because of the imperfections and inadequacies of language, its unskilful use and the inability of the human mind to foresee all the contingencies which will arise in the daily operation of the law. For this reason it is sometimes said that statutes should declare principles and not go into detail.

If important legislation is to be stated effectively in general principles it can be done only after very careful consideration by the drafters of all questions of detail and the selection of such general language as is suited precisely to the development and application of the general principle to the numerous particular instances to which it will be applied. Otherwise, the act is not truly general; it is simply incomplete.

There is an impression in this country that the English

'Compare remarks of F. Vaughn Hawkins, Esq., reprinted in Thayer's Preliminary Treatise on Evidence, appendix C, p. 585.

Workmen's Compensation Act is a good example of a well drafted act which states only general principles. In the case of Lysons vs. Knowles,' Lord Davey, in rendering his opinion in the House of Lords, referred to the act of 1897 as an ordinary ill-drawn act," and said:

extra

The difficulty really arises from this-that the draftsman has apparently not worked out on paper into legislative language the scheme which he had in his head, and it looks very much as if the act had really been framed from notes of legislative intention and had not been expanded into the proper legislative language. Cases which have arisen, and cases which are likely to arise, appear not to have been contemplated, but apparently were supposed to be covered by the general language used in the act.

The English Compensation Act of 1897 was expressed in 121⁄2 printed pages; the amended act of 1906 required 24 pages, and in addition there are now more than 150 pages of statutory rules and regulations which have the force of law. Compensation, under the act of 1897, was based on "average weekly earnings" without any indication of the method of computing such earnings. This computation gave rise to so many difficulties in the cases which arose under the act that the drafters of the amended act of 1906 used nearly 400 additional words to explain the method of computing average earnings, a total of 400 words in the place of the 3 words in the original act. The German Insurance Code of 1911 represents a like expansion of the original laws."

The tendency to couch statutes in general terms and to leave details of their administration to executive discretion simply shifts to executive officers the burden of applying the general principle to a particular case. This puts off the difficulty but

184 L. T. R. 65, vol. 3, Workmen's Compensation Cases (Minton-Senhouse), P. I (1901).

*The act and rules are reprinted in the appendix to Ruegg's Employers' Liability and Workmen's Compensation (1910), pp. 688-868.

First Schedule, sec. I, b.

First Schedule, section 1, clauses (1) and (2).

5 See translation in Bulletin No. 96 of United States Bureau of Labor.

does not overcome it; for if the law is to be even reasonably clear, executive officers must draft the rules and regulations and prescribe the schedules, reports and records, provision for which has been omitted from the statute. In this country, however, because of the general impression that such rules and regulations, supplementing general statutes, represent an unconstitutional delegation of legislative power, it usually happens that the general principle is applied in hit-or-miss fashion to each particular case as it arises. The New York labor law requires good and sufficient ventilation" in factories.' No specific rules have been prescribed and the act is practically unenforceable.

Moreover, when a general statute is well drawn, the men who have worked out its provisions and selected the language in which to state them are in a better position to state the specific rules for the application of the act to particular instances than are administrative and judicial officers before whom it comes as a totally new and often unconsidered matter. The drafters of a workmen's compensation act, for example, if they have done their work well, ought to know whether free house rent received by an employe is to be included in the computation of his wages for the purpose of determining his compensation in case of injury, and if they fail to state in their act whether it is to be included or not, employers, employes, insurance companies and courts are going to spend a great deal of time in attempting to discover whether the legislature intended to include or exclude this item, and no one is ever going to know what the legislature did intend until some individuals have carried to the court of last resort a case involving the question, and then the chances are even that the court will guess wrong and that the intent of the legislature if it had been expressed would have been directly opposite. For example, take the Sherman anti-trust law, the meaning of which was in doubt for twenty years. There are many people who, if they had been placed in the position of the Supreme Court, would probably have guessed differently as to the Congressional intent.

1 New York Consolidated Laws, ch. 31, sec. 86.