The Great Myth That
Plain Language Is Not Pr ecise
Joseph Kimble
O ccasi onally , w hen y ou try to convert from legalese to plain
lan guage, someone will com e forw ard and assert that y ou m ade a
m istak e. You m issed som ething in the tr anslation. You inadvertently
chan ged th e substance.
N ever m in d th at tr anslatin g legalese — lik e translatin g a foreign
lan guage — is no easy matter. N ever mind that, despite the
difficu lties, good w riters have successfully revised countless legal
docum ents into pl ain — or plai ner — lan guage. N ever m ind th at
many of these docum ents have invol ved tough subjects lik e finan cial
disclosur e, corporate takeovers, and disability insurance, not to
m ention the Feder al R ul es of A ppellate Pr ocedu re, A rticle 9 of the
U niform C omm ercial C ode, and var iou s federal r egulations issued
since the Pr esidential M emorandum on plain lan guage. 1 N ever mind
that for every inadv er tent chan ge, y ou could prob ably identi fy
several am biguiti es or u ncertain ties in th e origin al docum ent. N ever
mind that the revised document will almost certainly be better —
clear er and more accurate — than the origin al. T he fact rem ains that
revising and clarify ing a legal docum ent alw ay s involves som e
judgm ent and som e risk.
1
See DIV . OF C O R P . FIN A NC E, U . S. SEC . & EX C H . C O M M ' N , BEFO RE & A FT ER P LA IN
EN GLISH EXA MPLES AND SA M PL E A N A LY SES (1998); L A W R EF OR M C O M M ' N O F
V IC T O R IA , PL A IN EN GL ISH A N D T H E L A W app. 2, PL A IN EN GL ISH R EWRITE —
T A KEO V ER S C O DE (1987); DAVID ST . L. KELLY & C H R IST O P H ER J. BA L M F O R D, L IFE
IN S. FED ' N O F A USTR ALIA , SIMPLIFYING DISABILITY IN C O M E INSU R AN C E DOCU MENTS
(1994); Bryan A . Garner, The Substance of Style in Federal Rules, C LARITY No. 42, Sept.
1998, at 15; Steven O. Weise, Plain English Com es to the Uniform Com m ercial Code,
C LARITY No. 42, Sept. 1998, at 20; New Source Performance Standards for New Small
Municipal Waste Combustion U nits, 64 Fed. R eg. 47,275 (1999) (to be codified at 40
C.F.R. pt. 60) (proposed Aug. 30, 1999); Leasing of Solid Minerals Other Than Coal and
Oil Share, 43 C.F.R . pt. 3500 (1999)
110
The Scr ibes Jou rn al of Leg al Writin g
1998–2000
But the ri sk is wo rth it, and w riters should n ot be dissuaded.
O therw ise, the legal profession w ill never start to level the mountain
of bad forms and models that we have created. W e'l l be stuck w ith
the enorm ous inefficiencies of traditional style and the frustration it
causes. 2 C han ge is har d, but chan ge has to com e.
Let m e offer w hat I think is a perfect object lesson — a little story
from M ichigan.
The St or y in Br ief
The Michig an Bar Jo ur n al has a lon g-stan di ng col um n called “Plain
Lan guage,” w hich I happen to edit. In the October 1999 column, an
experi enced corporate attorney , David Daly , w rote an art icle cal led
“Taming the C ontr act C lause From H ell.”3 Daly undertook to
straighten out a m utual -in dem ni fication clause, on e in w hi ch each
party indem nifies the other. (I' ll give y ou th e fu ll clau se in a m inu te.)
A nother attorn ey then w rote a letter to th e edito r of the Bar
Jo ur n al; 4 th e letter po inted to th ree po ssible “errors” in Daly ' s
revised version.
Fir st, the revi sed version said that if the indemnified party is sued
and the indemn ify ing party assum es the defense, the indemnifying
party “m ay select counsel satisfactory to the other party.” The
origin al clause said that “the indemnifyin g party shall be entitled . .
. to assum e the defense thereof w ith counsel satisfactory to such
in dem ni fied party .” H m m . . . not m uch difference. But presum ably
2
See Joseph Kimble, An sw erin g t he C ri ti c s o f Pl ai n La n g u ag e, 5 SC RIBES J. L EGAL W R IT IN G
51, 62-65, 69-71 (1994-1995) (citing 15 studi es show ing that plain language impr oves
readers' comprehension of legal documents); Writin g fo r Do ll ar s, Writin g to Please , 6
SC RIBES J. L EGAL W RITING 1, 7-31 (1996-1997) (summ arizing 25 studies showing that
plain language saves time and money and i s strongly preferred by readers).
3
David T. Daly, Ta m i n g t he C o n tr ac t Clause from Hell: A Case Study, 78 M IC H . B.J. 1155
(1999).
4
Publ ished in “Opinion & Dissent,” 79 M IC H . B.J. 150 (2000) (letter originally dated
November 1999, w hen I first saw it and w rote this article).
1998–2000
The Great Myth
111
the intended sense is that the counsel m u st be satisfactory to the
other party . T hat could hav e been m ade clearer in bo th ver sions.
Second, th e r evised ver sion di d not say that until assuming the
defense, th e indem nify ing par ty m ust pay th e indem nified par ty ' s
legal fees. T he original said: “after notice . . . to assu m e the defense
th er eof, the indem ni fy in g par ty shall not be li able to such
indem ni fied par ty . . . for an y fees of other coun sel or an y oth er
ex pen ses, in each case subsequen tly incur red by such indem ni fied
party . . . .” N ow w hat does that mean — “in each case subsequently
incu rred”? (A nd why the comm a after e xp e n s e s?) O m it in eac h case
and y ou' ll pr obably get the in tended m eanin g. So that di vidi ng point,
w hen the indemnify ing par ty assum es the defense, shoul d have been
explicitl y stated in th e revised version.
T hir d, the origin al clause said that after assuming the defense, “the
indemn ify ing party shall have no liability w ith respect to any
com prom ise or settlement th ereof effected w ithou t its consent.”
W ithout going in to the reasons wh y , Daly dropped th at provi sion as
implicit in assuming the defense; and he added a comparable
provision for th e i n d e m n i f ie d party. H e might have explained the
chan ge. (A half-decent editor would have noticed and queried, but I
didn' t.)
A lth ough these three points in the letter to th e Bar Jo ur n al cannot
really be counted as errors, the points w ere w ell tak en. The troubling
part of the letter, though, w as this senten ce: “A s w ritten , the turgid,
repetitive and (nearly ) unreadable original has a paramount
advan tage over the concise, clearer version: it gets the int ended legal
relations right.” I hope no one w ill m ak e too mu ch of statements like
this. Please don' t conclude that a legal w ri ter has to ch oose betw een
precision and pl ain language — that legalese has the advantage of
being more precise, and plain language is less lik ely to get the
substance right. T hat' s just not true. In fact, it' s the biggest m y th of
all.
112
The Scr ibes Jou rn al of Leg al Writin g
1998–2000
A Look at the O r iginal C lau se
Let's back up and review this w ondrous clause. (The lines are
nu m bered so I can refer to them later.)
8. Indemnification
...
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
(c) Promptly after recei pt by an indem nified party under Section 1(g),
8(a) or 8(b) hereof of notice of the commencement of any action, such
indemni fied party shall, if a claim in respect thereto is to be made against
an indemnify ing party under such Section, give notice to the indemnify ing party of the commencement thereof, but the failure so to notify the
indemnifying party shall not relieve it of any liability that it may have to
any indemnified party except to the extent the indemnifying party
demonstrates that the defense of such action is prejudiced thereby . If any
such action shall be brought against an indemnified party and it shall give
notice to the indemnify ing party of the commencement thereof, the
indemnify ing party shall be entitled to participate therein and, to the
extent that it shall w ish, to assume the defense thereof w ith counsel
satisfactory to such indemnified party and, after notice from the indemnifying party to such indemnified party of its election so to assume the
defense thereof, the indemnify ing party shall not be liable to such indemnified party under such Secti on for any fees of other counsel or any other
expenses, in each case subsequently incurred by such indemnified party in
connection w ith the defense thereof, other than reasonable costs of
investigation. If an indemnify ing party assumes the defense of such an
action, (i) no compromise or settlement thereof may be effected by the
indemnify ing party w ithout the indemnified party' s consent (which shall
not be unreasonably withheld) and (ii) the indemnify ing party shall have
no liabi lity with respect to any compr omi se or settlement thereof effected
without its consent (w hich shall not be unr easonably withheld). If notice
is given to an indemnifying party of the commencement of any action
and it does not, within ten day s after the indemnified party' s notice is
given, give notice to the indemnified party of its election to assume the
defense thereof, the indemnify ing party shall be bound by any determination made in such acti on or any compromise or settlement thereof
effected by the indemnified party.
1998–2000
The Great Myth
113
In his article, David Daly sum marized w hy this thing is so poorly
drafted:
• The sentences don't begin w ith the m ain , or indepen dent, clause.
• T he sentences are too long.
• It u ses too m an y w ords.
• It fails to br eak th e m ater ial do w n into subpar ts.
True enough, but there's more. For all its supposed accuracy and
precision, the clau se is full of little h oles that th e den se surface hides.
One: shall is m isused throu ghout. Lawy ers are uneducable on shall,
and w e shou ld give it up. Com mentators and experts agree that it
should be used to im pose a duty . 5 It m eans “has a duty to.”
Essentially , if y ou can substitu te m u st, th en th e shall is correct. So
w hy not just use m u st to begi n w ith?
A t any rate, 8 of the 11 shalls in the indemnification clause are
m isused: the verb should be in the present tense. In lin e 11, for
in stance, it should be i s e n t i tl e d. Luckily , none of th e m isuses creates
an am bi gu ity , but our pr ofessional m ishan dl ing of shall betray s us in
more serious, problematic w ay s. You can see for y ourself by checking
W o r d s a n d Ph r a se s — 93 pages and over 1,200 cases dealing with shall.
T w o: in lin e 3, w hat does thereto refer to? The action? One of the
sections? T his ty pifies the pseudo-precision of our belov ed antique
jargon — w ords lik e thereto and herein and su c h.
5
BA R BA R A C HILD, DRA FTING L EGAL DOCU MENTS 204-06, 383-84 (2d ed. 1992); R EED
DIC KERSO N , T H E FU N D A MEN T A LS OF L EGAL DRA FTING §§ 6.7, 9.4 (2d ed. 1986);
BR YA N A. GA R N ER , A DIC T IO N A R Y O F M O D ER N L EGAL U SAGE 940 (2d ed. 1995);
Joseph Kimble, The Man y Misuses o f Shall, 3 SC RIBES J OU RNA L L. W R IT IN G 61, 64 (1992).
114
The Scr ibes Jou rn al of Leg al Writin g
1998–2000
T hr ee: in lines 4 an d 16, w hat do es su c h Se ctio n refer to? Does it
refer to one section in particular or to any one of th e three sections?
Four: lines 9–11 seem to say that th e indem nify ing par ty m ay
participate in the defense only if the indem ni fied par ty gives notice.
But w hy should that right depend on whether notice is given?
Five: in line 11, the an d shoul d be o r , right? That is, the
in dem nify ing party can participate in the defense without assum ing
the defense.
Six: lines 10–12 say that “the indem nify ing party shall be
entitl ed . . . , to th e exten t th at it shal l w ish, to assu m e the defense
thereof . . . .” Does this mean that th e indem nify ing par ty m ay
som ehow assum e part, but n ot all, of th e defen se?
Seven: in line 17, the i n e a c h c a se su b se q u e n t l y i n c u r r e d shoul d be in
this case, right? Better y et, omit in each case.
Eight: in lines 20, 23, and 29, w hat' s the difference betw een
“comprom ise” and “settlement”? Is this just another legal doublet, like
n ull an d v oid ? O r do y ou need bo th term s?6
N ine: lin es 26–27 say “w ith in ten day s after the indem nified party ' s
notice [of the action] is given.” Shouldn't that be ten day s after notice
is r e c e i v e d ?
Ten: lines 24–30 set a practical limit of ten day s on giving notice to
assu m e the defense. Is there any time limit on when the indemnify ing
party can decide to participate in the defense? A pparently not.
There m ay be m ore questions, but that's enough to bring home the
point: w hen y ou redraft in plain lan guage, y ou inevit ably un cover
gaps an d uncertai nties in legal istic wr iting. Th e fog lifts, the drizzle
en ds, and the light shines through. So I' ll say it again : plain langu age
is usually m o r e pr ecise than tradit ion al legal sty le. 7 The imprecisions
of legalese are just harder to spot.
6
See Garner, supra note 5, at 15 (noting a difference between co m pro m ise and se tt le m e n t,
but also noting that they “have been used with a variety of meanings and even
synony mously ”).
7
Kimble, Writing for Dollars, supra note 2, at 2.
1998–2000
The Great Myth
115
A not her Revised V er sion
David Daly ' s im prov ed plain-langu age version of the
indem nification clau se can easily be tw eaked to take into account the
com m en ts in th e letter to th e editor. Y ou could do it lik e this:
8. Indemnification
...
8.3 Legal Action Against Indemnified Party
(A ) Notice of the Action
A party that seek s indemni ty under § 1.7, 8.1, or 8.2 must
promptly give the other party notice of any legal action. But a
delay in notice does not relieve an indemnify ing party of any
liability to an indemnified party , except to the extent that the
indemnify ing party shows that the delay prejudiced the
defense of the action.
(B) Participating in or A ssuming the Defense
The indemnify ing party may participate in the defense at any
time. Or it may assume the defense by giving notice to the
other party . After assuming the defense, the indemnify ing
party:
(1) must select an attorney that is satisfactory to the other
party;
(2) is not liable to the other party for any later attorney ' s fees
or for any other later expenses that the other party incurs,
except for reasonable investigati on costs;
(3) must not compromise or settle the action without the
other party ' s consent (but the other party must not
unreasonably withhold its consent); and
(4) is not liable for any compromise or settlement m ade
without its consent. [Or om it this item as obvious?]
(C) Failing to A ssume the Defense
If the indemnify ing party fails to assume the defense within 10
days after receiving notice of the action, the indemnifying
party is bound by any determination made in the action or by
any compromise or settlement made by the other party.
116
The Scr ibes Jou rn al of Leg al Writin g
1998–2000
Letting G o of the Myth
The choice is not betw een precision and plain language. Plain
lan guage can be at least as precise — or as appropri ately vague — as
traditi onal legal writing. The choice is between perpetuating the
vices of four centuries and finally breaking free, betw een inertia and
advancem ent, between defending the indefensible and opening our
m inds.
Lawy ers continue to w rite in a sty le so im penetrable th at even
oth er lawy ers have trouble understanding it — as the debate over the
indem nification clause once again confirms. W hat w ould w e think
of engineers or doctors if they regular ly could n ot un derstand w hat
anoth er engineer or an other do ctor had w ritten ? Do y ou suppose
that the publi c' s im pression of law y ers is in an y w ay influenced by
ou r strange talk and even stranger w ritin g?
I k now that, in law , change is hard and progress is slow . But w hen
it comes to legal drafting — contracts, w ills and tru sts, statutes,
ordin ances — progress is glacial. W hy so? On e possible explanation
is that legal drafters are blindly overconfident. T hey believe that
becau se their forms have been around a long time, the forms must be
tried and true — a grossly exaggerated notion. 8 Or perhaps the
explanation is that legal drafters recogn ize that ever y one else' s
8
See DA VID M ELLINKOFF , T H E L A N GU A GE O F T H E L A W 278-79, 375 (1963) (“[T]he
formbooks . . . w ere decorated with decisions that had never passed on the language or
arrangement of the form. . . . [Moreover,] that vast storehouse of judicial definitions
known as Words and Phrases . . . is an impressive demonstration of lack of precision in
the language of the law. A nd this lack of precision is demonstrated by the ver y device
supposed to give law language its precision — precedent.”); C EN T R E F O R PL A IN L EGAL
L AN GUA GE, LA W W OR DS: 30 ESSA Y S O N L EGAL W ORDS & PH RA SES (Mark Duckw orth
& Ar thur Spyrou eds. 1995) (showing that a number of legal terms, lik e force m ajeure and
right, title , an d in terest, are not precise or not required by precedent); Mark Adler, Tried
an d Tested: The Myth Behind the Cliché , C LARITY No. 34, Jan. 1996, at 45 (showing that
a typically verbose repair clause in a lease is not required by precedent); Benson Barr et
al., Legalese and the Myth of Case Prece de n t, 64 M IC H . B.J. 1136 (1985) (finding that less
than 3% of the words in a real-estate sales contract had significant legal meaning based
on precedent).
1998–2000
The Great Myth
117
drafting is poor, but they can' t quite see the same deficiencies in their
ow n w ork . N ow that' s w hat y ou' d call a severely lim ited critical
sense. Y et Bry an G arn er, ou r leading auth ori ty on legal w ritin g,
offers this evidence from y ears of teaching:
In my C LE seminars on legal drafting, I routinely ask au dience mem bers to
answer tw o questions:
(1) W hat percentage of the legal drafting that y ou see is of a genuinely high
quality ?
(2) W hat percentage of legal drafter s wou ld claim to pr oduce high-quality
drafting?
A lt hou gh th er e' s some variation within any audience, the consensus is quite
predi ctable: the law y er s say th at 5% of th e legal draftin g they see is of a
genuinely high qual ity , an d th at 95% of the drafters wou ld claim to pr oduce
high-quali ty documents.
T her e' s a big gap there. It signals that there's still much consciousness-raising
needed w ith in the profession — especially on the tran sactional side. 9
Garn er asserts that, in gener al, w hi le li tigator s are ver y interested in
try ing to im prove their w ritin g, transactional law y ers are not.
In th e end, th e sham e for legal w ri tin g is not ju st that even
lawy ers have tro uble tr anslating it. T he shame is that legal w riting so
often and so undu ly needs translating because law y ers don't w rite in
plain language to begin w ith.
I give the last w ord on all this to the Law R eform C om m ission of
Victoria (Au stralia), w hich in the mi d-1980s produced a m onu m ental
four-volume study on plain language. Here is w hat the C om m ission
sai d about on e of its revisor y pr ojects:
If some detail has been missed, it could readily be included without affecting the
sty le of the plain English version. It would not be necessary to resort to the
convolut ed and repetitious sty le of the original . . . . Any errors in the plain
Engli sh version are the result of difficulties of translation, particularly
difficult ies in understanding the origin al version. Th ey are not inherent in plain
9
Bry an A. Garner, President's Letter, T H E SC R IV EN ER (Scribes — Am. Soc'y of Writers
on Legal Subjects, Fayetteville, Ark.), Winter 1998, at 1, 3.
118
The Scr ibes Jou rn al of Leg al Writin g
1998–2000
Engli sh itself. Ideally , of course, plain English should not in volve a translation.
It should be written from the beginning.
10
10
L A W R EF OR M C O M M ' N O F V IC T O R IA , PL A IN EN GL ISH A N D T H E L A W 49 (1987; repr.
1990).