Just Search page

Comprehending legal documents


by Felicity Kiernan (This article was originally published in Explain: the Newsletter of the Centre for Plain Legal Language, Issue 3, April 1995, pp. 3-4)

Comprehension of Legal Contracts by Non-Experts: Effectiveness of Plain Language Redrafting, Michael E J Masson & Mary Anne Waldron, Applied Cognitive Psychology, Vol. 8, pp. 67-85 (1994).

Plain Language Contracts and Reader Comprehension,
Michael E J Masson & Mary Anne Waldron, Proceedings: Just Language Conference 1992, The Plain Language Institute, Vancouver.

Masson and Waldron’s research answers some questions about the link between plain language and comprehension.  It also raises a number of important issues which need further investigation.

The research

Masson and Waldron wanted to answer three questions about plain language:


Masson and Waldron produced four versions of a set of legal documents.  These were:

1.  The original document

2.  The original document with archaic terms (eg hereinafter) removed or replaced

3.  Like version 2 but with the following changes:

- Long sentences broken up into shorter sentences

- Passive voice replaced with active voice

- Difficult words replaced with simpler words

- References to contracting parties (mortgagor, mortgagee) replaced with “you “ and “we”.

4.  Like version 3 but with all legal terms replaced or explained.

Testing

Masson and Waldron tested each document in three different ways:


Results

The results split the document into two groups: Versions 1 & 2 and Versions 3 & 4. Versions 1 & 2 (original and archaic terms removed) showed similar poor results in the comprehension testing, for paraphrasing, yes/no questions and long answer questions. Versions 3 & 4 (plain language and legal terms explained) had similar and significantly better results in the comprehension testing.

In other words, the change which most effectively increased reader comprehension was Version 3 with these features:


Simple word replacement, whether of archaic terms or of legal terms, had no significant effect on reader comprehension. Even though Versions 3 & 4 were significantly better than the first two versions, overall comprehension results were poor.

Replacing words is not enough

Masson and Waldron concluded that:


This study is a good empirical investigation of the link between plain language and readers’ comprehension.  It proves that replacing or explaining legal and archaic terms is not enough by itself to increase readers’ comprehension of legal documents.  Even better, the study shows that documents can be made significantly easier to comprehend with only four simple changes – those involved in Version 3.

Issues raised by study

Despite this, Masson and Waldron’s research raises two main issues which need further empirical treating to resolve. 

The first problem is with the second step in Masson and Waldron’s study, which produced Version 3.  This step involved four changes to the documents, while the other steps involved only one change.  Version 3 involved two changes to sentence structure (shorter sentences and active sentences) and two types of word replacement (simpler words and references to “you” and “we”).

The effect of grammatical changes

From the research provided by Masson and Waldron it is impossible to separate the effects of each of the four changes in Version 3.  Although they dismiss the effect of simple word replacement, the most effective change involved two types of word replacement.  It is useful to know that at least one of the changes in Version 3 is effective in increasing readers’ comprehension, but we still do not know which.

Further research needs to be done which separates the effects of word replacement and the effects of grammatical changes.  We need to test the effects of different types of grammatical changes, including passive to active voice, shortening sentences, removing embedded clauses and avoiding abstract subjects.

Plain language is not just words

The second issue raised by this research involves issues of design.  None of the versions produced by Masson and Waldron involved any changes to design features, such as typeface, justification, use of white space, structure of paragraphs, or the order in which information is presented.

We need to make sure that these design issues are recognised as an important part if the plain language process. We must also empirically test all the language and design changes that can make a document plain.  However, Masson and Waldron’s study is an important piece of research.  It shows that the language changes advocated by proponents of plain language have a significant effect on comprehension.  But we still need more research to isolate which changes have most effect.