Peter Butt, Faculty of Law, University of Sydney
Let me begin by congratulating the Law and Justice Foundation for its initiative in organising today’s seminar. The Foundation has for many years placed great emphasis on the importance of communicating the law in terms that non-lawyers can understand. About 10 years ago it funded the establishment at the University of Sydney of the Centre for Plain Legal Language, to help persuade lawyers, legislators and governments that legal information could and should be written in ways that ordinary people can understand. I was privileged to be one of the founding Directors of that Centre.
The large number of organisations represented here today is testimony to the Foundation’s enthusiasm for the subject. It’s also testimony to the importance of finding ways of communicating law more effectively than we are doing at the moment.
I don’t think it is difficult to justify the need for plain language law. Surely no-one can argue that the laws that bind us ought to be obscure. Surely no-one can argue that the Acts of Parliament that regulate us ought to be incomprehensible. Surely no-one can argue that the documents we sign – like contracts, agreement, wills, and the like – ought to be impossible to understand. To put it more positively, it surely must be better if the documents we sign are understandable. It must be better if, in a democracy, we could understand the laws that parliament passes. And yet we’ve all heard the adage: “ignorance of the law is no excuse” – well, it holds true even if the law is cast in such obscure language that the ordinary person simply can’t understand it.
No, there must be some good reason for using language that no-one can understand. The trouble is – the deeper you delve the harder it is to find a good reason; the harder it is to justify the dense, contorted, self-important style of writing we associate with legal documents and statutes.
I should begin by asking: what do we mean by “plain language law”, or (as it is often called) “plain legal language”?
“Plain language” is a term used by many. I suspect its meaning varies considerably from user to user. By “plain language” I mean language that is clear and idiomatic – for those who write in the English language, “plain language” is modern, standard English.
For some, “plain language” carries connotations of “dumbing down” the language – a kind of Dick-and-Jane style of writing that panders to the lowest common denominator. But this is a misunderstanding of the true nature of “plain language”, at least as practised by skillful proponents. In skilled hands, plain language uses the techniques of the very best writers, to produce prose that communicates directly and effectively with its intended audience.
“Plain language law” is simply the techniques of plain language applied in a legal context. It involves applying to legal documents and statutes those same techniques that good writers use in normal prose. It is effective writing, in a legal context.
To put “plain legal language” into context, we should contrast it with some examples of “traditional” legal drafting. Let me take just a few examples. We will see that they illustrate two main features – verbosity and undue technicality. They also ooze other characteristics: archaic language, illogical word order, complex grammatical structures, and sentences of excruciating length.
Leases are a prime example of all that is bad in traditional legal drafting. Suppose you want to impose on a tenant the obligation to repair the leased premises. You could write: “The tenant shall repair the premises” (or, preferably, “The tenant must repair the premises”). There is no doubt that, legally, this would suffice. “The premises” would be defined elsewhere in the lease. There would be no need list the various parts of the premises, because the term “the premises” would include all parts of the premises. And there would be no need to expand on the term “repair”, as it is an ordinary English word, whose meaning when used in leases has been elucidated by many judicial decisions. Yet compare that wording – “The tenant must repair the premises” – with the verbal excesses that appeared in the “repairing” covenant which gave rise to litigation in the English case of Ravenseft Properties Ltd v Davstone (Holdings) Ltd: 
“[The tenant shall] when where and so often as occasion requires well and sufficiently ... repair renew rebuild uphold support sustain maintain pave purge scour cleanse glaze empty amend and keep the premises and every part thereof ... and all floors walls columns roofs canopies lifts and escalators ... shafts stairways fences pavements forecourts drains sewers ducts flues conduits wires cables gutters soil and other pipes tanks cisterns pumps and other water and sanitary apparatus thereon with all needful and necessary amendments whatsoever ...”.
This is rampant verbosity, a verbosity which makes the clause far more difficult to read than its subject matter requires. Probably, the verbosity was prompted by a desire to be legally precise. If so, it failed, because the clause still ended up in court in a dispute over meaning. This demonstrates one of the great misconceptions of traditional legal drafting — that somehow a complex, traditional style is more precise than modern, plain language.
Of course, the problem is not confined to leases. Mortgages are just as bad — perhaps worse. One of Australia’s leading banks produced a standard mortgage which contained a sense-defying clause of 763 words — the clause contained 2 commas, 1 semi-colon, 3 sets of brackets, but no other punctuation.  As if not to be outdone by this Australian leviathan, a New Zealand bank’s standard guarantee form featured an entirely punctuation-less sentence of 1299 words; and the same document had an average clause-length of 330 words. As the great English conveyancer, Davidson, once wrote: the legal profession prefers “to seek safety in verbosity rather than in discrimination of language”. 
To some, these drafting feats may evoke admiration: after all, it takes skill to write a grammatically-perfect sentence of 1299 words. But for most readers the drafting serves only to bewilder. Sometimes it bewilders even the drafters themselves. In a 1992 Australian case, a bank tried to enforce a guarantee which a customer had signed. One of the customer’s defences was that certain clauses in the guarantee were meaningless. The guarantee form proved so tortuous that even the bank manager, when challenged in the witness box, had to admit that he could not understand some of the clauses; and it got worse – for, when challenged by the judge, nor could counsel for the bank. 
Despite the lawyers’ fondness for their traditional style of drafting, changes are afoot. The “plain language movement” is growing. The past 20 years have seen a clear trend in both public and private drafting towards a plain language style. The following are some illustrations of this growing trend:
I hope I have said enough to persuade you that traditional legal language is not all that its exponents make it out to be. I hope I have also persuaded you that the move towards using plain legal language is well-established. This brings me, then, to consider the real issue: why use plain legal language?
I think there are four basic reasons for using plain legal language:
Reason 1: that it is possible effectively to express legal concepts in plain language
This is the key to the whole plain language movement in law: that it is possible to express legal concepts in plain language, without loss of certainty and precision, even in areas of law that are complex. It is the assumption on which all plain language legal drafters rely. Without it, there would be no plain language movement in law.
The validity of the assumption is, I think, borne out by evidence from several different – and difficult – areas of law. One is corporations law. Much of our corporations law is now drafted in plain language – with (so far as I know) no litigation over meaning. This gives the lie, I think, to those who would say “complex law requires complex drafting”. Another example is taxation law. In a study back in 1991, the Victorian Law Reform Commission tested the comprehensibility of part of the then-current Australian income tax legislation. The result? To understand the legislation required 12 years of schooling plus 15 years of university — 27 years’ education in all.  But this is now changing. Australia now has tax law drafted in language that is brutally plain; New Zealand is following suit; and the UK government has established a tax rewrite program to introduce plain language taxation laws.
The validity of this assumption – that it is possible effectively to express legal concepts in plain language – is also borne out by experience with private legal documents. There is no evidence that plain language legal documents are more prone to litigation than conventionally-drafted ones. Indeed, the reverse seems to be the case. To illustrate: for more than 20 years, the NRMA has been using insurance policies that are exceedingly plain. Some time ago the company’s managing director went into print to say that their policies had been entirely free of litigation.  Now, a cynic might put this down more to good luck than good management – for eventually some plain language documents will end up in court; drafters are only human. But at least the evidence to date gives the lie to any argument that plain language documents are inherently prone to litigation.
This is not to say that writing legal documents in plain language is easy. Legal writing poses problems not usually found in other forms of writing. A leading issue is, what do we do with legal “terms of art”? Here there are some differences of approach. Some plain language proponents do their best to eliminate terms of art altogether — they find some other way of expressing the legal ideas inherent in the term. The danger with this, of course, is that legal precision can be lost unless the new words accurately capture the legal nuances of the original. This may require real skill, and a great deal of research. And so another way is to retain terms of art, but then to add an explanation of what the legal word or phrase means – a sort of “best of both worlds” approach.
Another problem is what to do with terms whose meanings have been judicially defined? Lawyers are afraid to dispense with those terms, in case they lose the benefit of judicial definition and risk the possibility of litigation over the meaning of new terms. However, we should not exaggerate the problem; the problem is not nearly as great as many opponents of plain legal language seem to imagine. Research shows that, in any given area of law, the number of legal terms which have been judicially-defined, is likely to be quite small. For example, studies in the United States of America show that the proportion of judicially-defined terms in standard form contracts for the sale of land may be as low as three percent.  The implication is that you can change the other 97% without losing the benefit of judicial exegesis. And some of the three per cent required judicial exposition for the very reason they were inherently uncertain – those terms it would be best to avoid altogether.
And of course nothing justifies using jargon for its own sake. Nothing justifies perpetuating linguistic eccentricities that serve only to enhance mystique, not legal effect. And yet lawyers still introduce documents with “whereas”. They “execute” them rather than sign them. They “demise” rather than lease. They require a tenant to “well and sufficiently repair” when “repair” will do. They declare something “null and void”, when “void” will do. They insist on “shall” when the rest of the community uses “must”.  And so on. None of these hallowed words and phrases is a true term of art. All can be simplified, and some can be discarded completely.
Reason 2: that plain legal language saves money
Here the evidence is unequivocal and overwhelming. Many studies show that plain language is more “efficient” and therefore saves money.
By “efficient”, I mean that plain language documents are easier to read and comprehend. Numerous organisations attest to saving substantial amounts of money by converting their documents into plain language. Insurance companies are a prime illustration: by rewriting documents into plain language, enquiries from customers about meaning are reduced; this allows the company to redeploy enquiry staff to other tasks. And by redrafting forms in plain language, error rates are reduced; this saves time and money for the company, and helps cut down frustration for the customer. Studies of other organisations – including government bodies – show similar results. To take a stark example: some years ago, British Post redrafted its redirection-of-mail forms. Before the re-draft, there was an 87% error rate when customers filled out the form. Royal Mail was spending over £10,000 a week to deal with complaints and to reprocess the incorrect forms. The new form reduced the error rate dramatically – so much so that Royal Mail saved £500,000 in just the next nine months. 
These efficiencies are not unique to lay readers. Lawyers, too, save time and money when documents are in plain legal language. Lawyers find plain language easier to read and digest, cutting down time and effort for them almost as much as for their clients. These efficiencies have been documented in a study by the Law Reform Commission of Victoria. In the study, lawyers read counterpart versions of the same statutes, one version written in plain legal language and the other in traditional legal language. The time taken to understand the plain language version was between one-third to one-half less than the time taken to understand the traditional version. 
Reason 3: that judges prefer plain language
This reason is perhaps not as important as the others, but studies have borne it out just the same. Opponents of plain language have been heard to argue that plain legal language is “dangerous” because judges don’t like it. The argument goes that documents and statutes must be drafted to be litigation-proof, and for this it is important to get the judge “on side”; and as judges prefer traditional styles of drafting, it is safer to stick to traditional styles of drafting. This argument seems (to me) rather facile; but even if we accept this view of judges, the evidence is that, given a preference, judges would choose plain language. Studies in the United States show that something like 80% of judges surveyed preferred pleadings to be in plain language rather than in the traditional, convoluted American style.  Interestingly, the same judges also thought that lawyers who drafted pleadings in plain language were better lawyers than those who stuck to traditional drafting. I don’t know of any similar studies outside of the USA, but I would imagine that similar results would follow in most jurisdictions.
Certainly, in recent years some English and Australian judges have shown an increased willingness to condemn from the bench legal drafting that is convoluted and unclear. Epithets heaped upon offending clauses have included “botched”,  “half-baked”,  “cobbled-together”,  “doubtful”,  “tortuous”,  “archaic”,  “incomprehensible legal gobbledegook”,  and “singularly inelegant”.  (If I may be allowed some free advertising: you will find these and other examples in Butt and Castle, Modern Legal Drafting, Cambridge University Press, 2001, chapter 2.)
To be even-handed here, some Australian judges have also been less than enthusiastic about plain language in law. For example, an appeal court judge from Victoria recently described certain provisions of the Corporations Law as being drafted “in the language of the pop songs”.  In this judge’s view, the quest for simplicity “pays the price of vulgarity and ends in obscurity”.  Another Australian appellate judge decried as “grotesque” the use of “must” in statutes, especially the phrase “must not”.  Yet another appellate judge, when criticising a clause in a plain language insurance policy, caricatured “plain English” as meaning “confused thought and split infinitives”,  as if it served no useful purpose at all. And another described a plain language insurance policy as “one of those new fangled plain English policies which is, accordingly, a little hard to construe”.  But these are isolated voices and the tide of time will pass them by. I suspect that most judges would accept that modern Australian statutes and documents, which are now increasingly being drafted in a plainer style, are far easier to read and apply than their traditionally-drafted forebears. Indeed, some appellate judges – more farsightedly – have made constructive comments, accepting change as inevitable and moving on to wrestle with the very real issue of the extent to which settled case law can be applied in interpreting plain English revisions of statutes and standard documents. 
Reason 4: that the public prefers plain legal language
This last reason is also borne out by evidence – not just anecdotal evidence, but empirical evidence from large-scale research. Non-lawyers prefer legal documents and statutes to be in plain language. Amongst the research is a Canadian survey,  which showed a widely-held public perception that lawyers care little about whether they communicate effectively with their clients. Lawyers, the public thinks, are pre-occupied with legal precision at the expense of clear communication – they are indifferent to whether their clients understand the documents they are asked to sign. Lawyers may think that they do care whether they communicate – but the public perception seems to be otherwise.
The evidence is clear. Members of the public – particularly those with no legal learning – prefer plain legal language. If they are clients, it gives them a better chance to understand the legal consequences of the documents they sign; if they are citizens, it gives them a better chance to understand the laws that bind them.
I said earlier that the plain language movement is now reasonably well-established. In its early days, it made assumptions about the benefits of plain legal language – without at that stage having verified the assumptions by empirical research. But now, about 20 years on, research has proved the assumptions to be correct. The evidence is overwhelming. Plain legal language brings substantial benefits to lawyers, to clients, and to citizens at large. It can be legally safe; it saves time and money; clients and citizens have a better chance of understanding it; and most judges prefer it. The evidence is all one way. I would suggest that there is no substantial reason to resist it.
This paper was presented at the Speaking Plainly seminar organised by the Law and Justice Foundation of NSW on 12 September 2002.
 .  1 All ER 929.
 . See Memorandum filed in New South Wales Registrar-General’s Office No V581852, clause 1(g).
 .Davidson’s Conveyancing Precedents, 3rd ed 1860, Vol 1, p 67.
 . Houlahan v Australian and New Zealand Banking Group Ltd (1992) 110 FLR 259.
 . These are particularly active in the United States, in California, Michigan, Texas and Pennsylvania. The Michigan and Texas Bar Association Journals publish regular plain language columns.
 . Examples include the New South Wales Law Society’s Standard Form Contract for the Sale of Land (latest edition, 2000); and the England and Wales Law Society’s Standard Conditions of Sale (1999), Standard Commercial Property Conditions (1999), and plain language business lease.
 . Adler, Clarity for Lawyers (1990, Law Society of England and Wales); Perrin, Better Writing for Lawyers (1990, Law Society of Upper Canada).
 . Victorian Law Reform Commission, Plain English and the Law (1987).
 . New Zealand Law Commission, A New Interpretation Act (1990).
 . Irish Law Commission, 1999.
 . Two examples are the Land Title Act 1994 (Qld); and the 1998 amendments to the Native Title Act 1993 (Cth).
 . Turnbull, “Clear Legislative Drafting: New Approaches in Australia” (1991) 11 Statute Law Review 161; Murphy, “Plain Language in a Legislative Drafting Office” (1995) Clarity (33) 3.
 MD Kirby, Book Review of Decisions, Decisions … a handbook for judicial decisions (by Justices Mailhot and Carnwath) (1999) 73 ALJ 290 at 292.
 . Australia, New Zealand, and the UK. The drafters came together for a conference in Auckland, NZ, in November 1996; for those interested in the thought processes of drafters, the published conference proceedings make fascinating reading.
 . Clyde, “Plain Language Turns the Corner: New SEC Rules for Prospectuses”, Clarity 42 (September 1998), pp 9-14.
 . Legal Profession Act 1987 (NSW), s 179(1).
 . Workplace Relations Act 1997 (Qld), s 348.
 . Industrial Relations Reform Act 1993 (Cth), s 17 (inserting a new provision, s 150A, into the Industrial Relations Act 1988 (Cth). Compare Lord Diplock’s lament in Merkur Island Shipping Corporation v Laughton  AC 570 at 612, that the law “should be expressed in terms that can easily be understood by those who have to apply it even at shop floor level”.
 . For example, Consumer Credit (NSW) Code, s 162.
 . See Garner’s Guidelines for Drafting and Editing Court Rules (undated).
 See Australian Financial Review, 27 September 1991, p 19.
 Neville King, “An Experience with Plain English” (1985) 61 Current Affairs Bulletin (January), p 21. To similar effect, in its submissions to the Australian Parliamentary Inquiry into Commonwealth Legislative and Legal Drafting (18 September, 1992), the same company (the NRMA) stated: “The NRMA has not experienced any adverse court decisions by reason of the ‘Plain English’ and subsequent ‘user friendly’ documents.”
 .Barr, Hathaway, Omichinski and Pratt, “Legalese and the Myth of Case Precedent” (1985) 64 Michigan Bar Journal 1136.
 . On the use of “must” instead of “shall”, see correspondence in the Australian Law Journal: (1989) 63 ALJ 75-78, 522-525, 726-728; (1990) 64 ALJ 168-169. For a more detailed survey, see Kimble, “The many misuses of Shall” (1992) 3 Scribes Journal of Legal Writing, 61-77. At least one Australian case has expressly recognised that “must” is quite sufficient to impose an obligation: South Australian Housing Trust v Development Assessment Commission (1994) 63 SASR 35 at 38.
 This and similar examples are given in Kimble, “Writing for Dollars, Writing to Please” (1996-1997) 6 Scribes Journal of Legal Writing 1. For other studies, see Kimble, “Plain English: A Charter for Clear Writing” (1992) 9 Thomas M Cooley Law Rev 1 at 25-26; Mills and Duckworth, The Gains from Clarity (Centre for Microeconomic Policy Analysis and Centre for Plain Legal Language, University of Sydney: 1996).
 . Eagleson, “Plain English - A Boon for Lawyers”  The Second Draft (Legal Writing Institute) p 12.
 Harrington and Kimble, “Survey: Plain English Wins Every Which Way” (1987) 66 Michigan Bar Journal 1024; Kimble and Prokop, “Strike Three for Legalese” (1990) 69 Michigan Bar Journal 418; Child, “Language Preferences of Judges and Lawyers: A Florida Survey” (1990) 64 Florida Bar Journal 32. For articles by judges themselves, see Mester, “Plain English for Judges” (1983) 62 Michigan Bar Journal 978; Cohn, “Effective Brief Writing: One Judge’s Observations” (1983) 62 Michigan Bar Journal 987.
 . Re Gulbenkian’s Settlement  AC 508 at 517 (Lord Reid).
 Alghussein Establishment v Eton College  1 WLR 587, HL.
 . The Alexion Hope  1 Lloyd’s Rep 311 at 320, CA (Purchas LJ).
 . Overseas Union Insurance Ltd v AA Mutual International Insurance Co Ltd  2 Lloyd’s Rep 63 at 69 (Evans J).
 . London Regional Transport v Wimpey Group Services Ltd (1987) 2 EGLR 41 (Hoffmann J).
 . Trafalgar House Construction (Regions) Ltd v General Surety & Guarantee Co Ltd  3 WLR 204 at 212 (Lord Jauncey of Tullichettle).
 . Houlahan v Australian and New Zealand Banking Group Ltd (1992) 110 FLR 259.
 . NSW Rifle Association v Commonwealth of Australia, unreported, New South Wales Court of Appeal, 15 August 1997, Powell JA.
 . GM & AM Pearce and Co Pty Ltd v RGM Australia Pty Ltd (1998) 116 ACLC 429 at 432 (Callaway JA).
 . Ibid.
 . Hallwood Corporation Ltd v Roads Corporation  2 VR 439 at 445-446 (Tadgell JA). For criticisms of the judge’s comments, see Eagleson, “Plain English: Changing the lawyer’s image and goals”, a paper delivered to Literature and the Law Seminar, Perth, Australia, 16 May 1998; extracts published in (1998) Clarity 42, p 34.
 . NRMA Insurance Ltd v Collier (1996) 9 ANZ Insurance Cases 76,717 at 76,721 (Meagher JA).
 . Re Network Welding Pty Ltd (in liquidation) (No 2)  NSWSC 809 at para 3 (unreported, 28 August 2001).
 . See, for example, Justice G Hill, “A Judicial Perspective on Tax Law Reform” (1998) 72 Australian Law Journal 685; Justice K Lindgren, “Interpretation of the Income Tax Assessment Act 1997” (1999) 73 Australian Law Journal 425; Justice D Mahoney, “A Judge’s Attitude to Plain Language”  New South Wales Law Society Journal (September) 52.
 Survey carried out by the Plain Language Institute of British Columbia: see the Institute’s Preliminary Report, “Critical Opinions: The Public’s View of Legal Documents” (1992).