By Jack Stark
One salient recent development in statutory drafting is the spread of the Plain Language School’s influence. For some reason, the leading proponents of this school have focused intently on statutes. In response to this attention, many statutory drafters have accepted the school and use its precepts. For that reason, it is useful to examine and evaluate Plain Language.
One basic assumption of the Plain Language School as it applies to drafting is that laypeople frequently read statutes. They do not. Some drafters claim the number of hits on their Internet site statute pages proves many laypeople use them. These claimants, however, do not know (1) how many have accessed their Internet statutes because most users hit on more than one statute or (2) who the users are. If the site is updated more than once a session, has links to cross-references or has a search function, it is a better research tool than the printed statute books and is being used by the true audience for statutes—lawyers, judges and administrators. Another way to demonstrate that laypeople do not frequently read statutes is to ask yourself whether you consulted the Internal Revenue Code when you prepared your recent income taxes. If you, a professional reader and writer of statutes, did not, why do you think laypeople did? The main reason you did not is that the form and instructions probably are sufficient. If not, Publication 17 is available and, if that is not enough, the Internal Revenue Service has produced many narrowly focused publications on federal income taxes.
Another basic assumption of the Plain Language School is that citizens have a right to read simplistic statutes. Rather, they have a right to access the law. If one wants citizens to have access to the law, one ought to write publications such as those of the Internal Revenue Service and to encourage executive agencies to do so. That is the course to take to help laypeople. Pamphlet writers acknowledge that statutes do not represent the entire law, a mistake Plain Language proponents often make. Statutes are interpreted by rules or regulations and legal opinions and relate to other statutes. Hence, there is a need for explanatory publications.
By now, it should be clear that I have some doubts about Plain Language. To test my reactions, we need an example of a Plain Language revision of an existing statute or regulation. The proof is in the pudding. I went to a randomly chosen Plain Language Internet site, randomly chose a Plain Language revision, then took the first relatively short, revised regulation I found. To me, it seems representative. If you disagree, you might try the same method and make a meticulous and objective evaluation or read Euan Sutherland’s devastation of a Plain Language revision of an act that he had drafted for the English Parliament. It can be found at 14 Statute Law Review, Issue 3 pp. 163-170 (1993). Another useful article is Brian Hunt’s general discussion of Plain Language at Statute Law Review 23, Issue 1, pp. 24-46 (2002).
Here is the sample I found. The regulation appears first, then the revision:
12 CFR Sec.602.272 Aggregating requests. A requester may not file multiple requests at the same time, each seeking portions of a document or documents solely in order to avoid payment of fees. When the Farm Credit Administration reasonably believes that a requester, or a group of requesters acting in concert, is attempting to break a request down into a series of requests for the purpose of evading the assessment of fees, the Farm Credit Administration may aggregate any such requests and charge accordingly. One element to be considered in determining whether a belief would be reasonable is the time period over which the requests have occurred.
12 CFR 602.16 Combining requests. You may not avoid paying fees by filing multiple requests at the same time. When FCA reasonably believes that you, alone or with others, are breaking down a request into a series of requests to avoid fees, we will combine the requests and charge accordingly. We will assume that multiple requests within a 30-day have been made to avoid fees.
The revision is shorter and easier to read, but let’s see whether it has distorted the meaning of the original version. In the title and the text, “aggregating” and “aggregate” have been changed to “combining” and “combine.” The meaning has been changed. The requests will be added up: “aggregated.” They will not be blended together to make a big request: “combined.” “Seeking portions of a document or documents” has been eliminated; the rules now apply to any request. “Solely” has been eliminated, allowing other causes such as forgetting that a request has already been made and that the agency erred. “Acting in concert” has been replaced by “with others,” which includes requests made at the same time by chance and requests with several names on them. “Series,” which means any number of related entities, has been replaced with “multiple,” which means many, not more than one. That means the original version of the regulation applies to two requests or more, but the revision applies only to many requests— who knows what number that means? “Is attempting to break a request down” has been changed to “are breaking down a request,” thus transforming an attempt into a completed act. “May aggregate” has been changed to “will combine,” which is a change from a permission to a requirement. “One element to be considered” has been transformed into “we will assume,” making an automatic assumption, not merely a consideration. Motive, circumstances and consequences become irrelevant. Stating that the time period is a consideration has been changed inexplicably to a 30-day period. Finally, the reviser missed the contradiction in the original, which refers both to requests filed at the same time and requests filed during an indefinite period. The revision keeps the contradiction but substitutes a 30-day period for the indefinite time span. The failure to correct the contradiction indicates that the reviser merely replaced words and phrases that are not Plain Language with words and phrases that are Plain Language. That is, he or she acted mechanically rather than thoughtfully. That is the foundational error Plain Language makes. It also causes 10 errors in four and a half lines of prose. My doubts about Plain Language have been confirmed.
How can a method of drafting that generates so many errors have been propounded, much less adopted? One reason for the propounding is that most of its public advocates are not professional drafters but academics and others who may never have drafted a bill. Another reason is that the advocates believe language is uniform, so the same set of rules will be universally applicable—the same rules can be applied to telephone books and Elizabethan sonnets. A better assumption is Ludwig Wittgenstein’s belief, as developed in Philosophical Investigations, that many language-games exist, each with its own conventions, functions and tactics. One such game—and an extremely complex one—is statutory drafting. In using Plain Language, it is much easier to be clear—one need only follow a few simple rules—than it is to be accurate, which requires considerable knowledge of statutes and the tactics that allow one to draft accurately. Plain Language drafters also mistakenly think they are abetting democracy; whereas, they are distorting statutes and not writing publications that will really allow citizens access to the law.
Plain Language distorts statutes because its theoretical underpinning is shot through with fallacies. The first is that it is more important to be clear (easy to read) than to be accurate (to effect the intent of the bill’s requester as amended). However, an accurate draft that is slightly difficult to read is a success, but a draft that is easy to read but inaccurate is a failure. This problem cannot be solved by claiming that a drafter can simultaneously be accurate and clear. The two goals are based on contrasting assumptions and tactics. It also cannot be solved by thinking a drafter can first write an accurate draft and then make it clear. The example above demonstrates the ineffectiveness of that approach. It also cannot be solved by assuming accuracy and concentrating on clarity. Accuracy is difficult to achieve.
Another fallacy is that Plain Language rules are sufficient for effective drafting. There is no agreed-upon list of Plain Language rules, although there is some consistency. Typically, there are lists of 10 or 12 rules, far too few for an enterprise as difficult as statutory drafting. Further, many rules make no sense. One rule, illustrated in the test case above, is to address “you.” It is not clear who “you” is, but it probably the person reading the statute or regulation. Imagine a statute that decrees, “You shall pay the shipping charges.” If the buyer is reading the statute he or she must pay the charges; if the seller is reading it, however, he or she must pay the charges. That state of affairs is unacceptable. It would be much better to specify explicitly who is to pay.
A third fallacy is the command that short sentences should be used. This is another method to over-simplify statutes; it also makes no sense. Moderately long sentences are needed to connect ideas, rather than leaving it to the reader to make the connection. For example, the three behavioral directives that dominate statutes are forbidding, authorizing and requiring. They are accompanied by conditions (stating those that make the behavioral directives functional) and consequences (for following or not following a behavioral directive). Fairly long sentences are needed to firmly connect a behavioral directive with the pertinent conditions or consequences. One way to create short sentences is to eliminate apparent redundancies. An example given by a Plain Language advocate uses the phrase “give, bequeath and devise.” At first glance, eliminating “bequeath and devise,” as the advocate recommends, seems both to reduce a sentence’s length and to eliminate legal jargon laypeople do not use. However, the three words have different meanings. In the language-game of legal writing, “give” denotes making a gift from one live person to another and being a transfer that fulfills the requirements developed in case law, such as having a donative intent and acceptance. “Bequeath” denotes transfer of personal property at death, and “devise” denotes transfer of real property at death. All three words are needed.
The rule to write about only one issue in a paragraph is also nonsensical. There are no issues (subjects on which there is disagreement) in the statutes; a better term is “topic”. Also there are more components in addition to paragraphs. The building blocks of statutes are statutory units such as sections and chapters. In addition, limiting statutory units to one topic is too restrictive, unless the drafter writes only one-sentence units, which would result in a collection of disconnected statutory units.
Two additional, less important flaws exist in Plain Language theory. One is the de-emphasis on substance in favor of form: following the simplistic rules. The other is a misdirected focus. Three possible foci can be used in an analysis of writing documents: the text, the reader and the writer. Plain Language focuses on the reader. It would be more useful to focus generating drafting tactics that will enable the writer to think and draft more accurately. An example is being aware of the five functions used in most statutes.
The most damaging Plain Language rule is to write only words that are commonly used by laypeople in ordinary speaking and writing. As one Plain Language Internet sites puts it, “Plain language is language that everyone in your audience can easily understand” (www. plainlanguagenetwork.org). For one thing, some legal terms have no Plain Language synonyms. ”Tort,” for example takes its meaning from the thousands of tort cases. Another problem with this rule is that ordinary language often is inaccurate. An example of this is the use of “mortgage,” which means a security agreement given by a borrower to a lender. This definition means all the following expressions are misnomers: “mortgage lender,” “mortgage rate,” “mortgage interest,” “paying off the mortgage” and “getting a mortgage.” In each instance, the proper phrase is “home loan.” Also, the lender is the mortgagee, and the borrower is the mortgagor. It is foolish to follow the crowd on this topic. Using simple words also reduces the intellectual level of statutes. I would be embarrassed to admit that my job is to write dumbed down statutes. Drafters are trained professionals, and should write like trained professionals, not like high school students. Finally, and most important, using simple words generates many errors, as seen in the test case.
In short, Plain Language is a disaster because it generates many errors, because it is illogical and ignores the statutory drafting language-game. Drafters who use it may cause irreparable damage to their statutes and thus may cause irreparable damage to their states. They need to rethink their methods.