Chapter 25

Analysing Ambiguity

Introduction
1. Ambiguity
Identifying Meanings
Classification of Ambiguity
Scope of Ambiguity
Prevalence of Ambiguity
Relevance of Ambiguity
2. Limits to Interpretation
General Limits: Language
Specific Limits: Ambiguity
Commentary

Introduction
Model for Interpretation
To assist readers to understand this chapter better and to appreciate its significance it is
necessary to introduce them briefly to the model for interpretation. Since this model is
developed later in this book a summary will suffice here.1 The model has three steps.
These are Step 1: Options, Step 2: Reasons and Step 3: Decision.

Step 1: Options
This step identifies the options before the court. These consist of the various meanings
of the ambiguous provision and the effect that each meaning will cause if chosen as
the legally correct meaning. These meanings are designated Meanings 1–n and their
effects as effects as Effects 1–n. Meaning 1 causes Effect 1, Meaning 2 causes Effect 2
and so on.

Step 2: Reasons
This step formulates reasons for and against each meaning of the ambiguous provision.
The aim of the reasoning process is to identify the meaning whose effect is the best. In
our analysis, this is the effect with the highest net benefit. The meaning that causes this
effect is the best meaning.

Step 3: Decision
Step 3 is a formality since it flows from Step 2. The reasoning process in Step 2
identified the best meaning. In Step 3 the court formally makes the decision that
decrees this meaning to be the legally correct meaning of the ambiguous provision.
Obviously a lawyer who is advising a client does not decide the correct meaning as the
court does. Instead they try to predict the meaning that the court will choose.

1. Chapter 30 Model for Forming Law

381
 382 Chapter 25 Analysing Ambiguity

Analysing Language
There are two reasons for analysing language when working with law. Both are
connected to interpreting law. Analysing language is necessary for interpretation in
that a vital part of the method for interpreting law is to identify precisely the
ambiguity, including the meanings that constitute the ambiguity, in the provision in
statute or common law that makes interpretation necessary. Analysing language is
necessary for determining the limits of interpretation – in any given case it is necessary
to know what meanings are arguable and what meanings are not.

1. Ambiguity
Ambiguity creates the need for interpretation and yields the range of meanings that
confront a court. (These are represented as Meanings 1–n in our model). While there
has been some consideration of ambiguity by legal writers,2 given that ambiguity is so
fundamental to understanding interpretation there has been surprisingly little detailed
analysis of it. While most theories of interpretation address ambiguity in some way,
the treatment is often cursory rather than detailed. Conspicuously absent so far is a
classification of ambiguity to assist in detecting, understanding and resolving it.

This discussion seeks to rectify these problems at least in part. To start, it highlights
the relevance of ambiguity by explaining why it is so important to identify the various
meanings of an ambiguous provision when interpreting law.

Discussion then turns to techniques that lawyers can use to identify ambiguity.
Ambiguity comes in several forms so one technique involves using a classification
system for ambiguity as a check list and a guide. Since this classification is so
important, this chapter pays it considerable attention.

Then the discussion confronts two controversies. One concerns the scope of
ambiguity, which entails resolving a dispute as to whether ambiguity should be given a
wide or a narrow meaning. The text argues for a wide meaning on functional grounds.
(Prior discussion of identifying meanings and classifying ambiguity was premised on
this wide view of the scope of ambiguity.)

The second controversy concerns the prevalence of ambiguity. Is language fraught

with ambiguity on a pandemic scale, or is it something that happens sometimes? Our
argument is that it happens sometimes rather than most of the time.

Identifying Meanings
Introduction
When interpreting law it is necessary to analyse the ambiguous provision to identify
the meanings that constitute the ambiguity. It is not enough just to sense that the
provision is ambiguous. Step 1 in the model for interpreting law directs those who
interpret law to uncover these various meanings and then to frame them in the clearest

2. Commentary 25.1.
 Chapter 25 Analysing Ambiguity 383

possible way. In short, Step 1 requires precise identification of the meanings that
constitute and cause the ambiguity.

Because precise identification of the meanings is so crucial, lawyers need to develop a

technique for performing the task. Three techniques are explained here – using a
dictionary, reusing the words in another context and consulting a catalogue that
contains a classification of ambiguity.

Dictionary
A simple and highly useful means of ascertaining possible meanings is to look up the
word or words in a dictionary. A standard English dictionary will at least give the
broad sense of a word by indicating a commonly accepted range of meanings. In this
regard, dictionary definitions are a useful, almost necessary starting point for
interpretation. Be aware, though, of the limitations of a dictionary – for example, no
dictionary is perfect and words shimmy about. Moreover, some of the forms of
ambiguity would not ordinarily be captured or completely captured in the dictionary
definition of a term.3

Obviously a dictionary merely indicates possible or permissible meanings of word and

even than not infallibly. A dictionary describes only ‘how a word can be used’ but
emphatically does not describe how it is ordinarily used or which meaning should
prevail in a particular context.4

Reuse Words
Another method is to take the words under scrutiny and use them in another context.
This gives a fresh look at the provision. It also enables you to make a contrast between
two settings where the provision is used and in this way may enlighten you as to the
ambiguity and even as to its possible resolution.

There is a good illustration in Corkery v Carpenter.5 There, Corkery had been charged
with being drunk while in charge of a ‘carriage’ because he was pedalling his bicycle
while under the influence of alcohol. The issue, therefore, was whether a bicycle was a
‘carriage’. To show popular usage, counsel recited a popular music hall song. As far as
relevant it went: ‘Daisy, Daisy, give me your answer true, I’m half crazy, all for the
love of you. It won’t be stylish marriage, I can’t afford a carriage, but you’ll look
sweet upon the seat of a bicycle built for two’.6

Through the intensity of this marriage proposal readers can see how the song
illustrates that in popular usage the word ‘carriage’ is taken not to include a bicycle.
Whether this was the legally correct position was precisely the issue in this case.
Unfortunately in the case, the rhetoric of romance did not prevail and Corkery, the

3. See Christopher Enright Legal Method, Chapter 14 Classifying Meanings.

4. Smith v United States 53 113 SCt 2050, 2061 (1993) per Justice Scalia
5. Corkery v Carpenter [1951] 1 KB 102
6. Corkery v Carpenter [1951] 1 KB 102, p 103
 384 Chapter 25 Analysing Ambiguity

inebriated pedaller, was convicted. The court held that the core meaning of ‘carriage’
was something that carries things or people – in other words a form of conveyance.
Since a bicycle carried people it was a carriage.

Classification of Ambiguity
A catalogue of ambiguity is an aid to identifying ambiguity because it classifies
ambiguity. It does this by setting out various categories and subcategories of
ambiguity. A lawyer who is interpreting law can use these as a check list. These forms
of ambiguity are discussed in detail in another text and are summarised below.7 Some
of these move beyond the entries in a dictionary so that they include other ways in
which words can have two or more meanings.8

Classification of Ambiguity
Introduction
As just discussed, one of the ways to identify ambiguity is to use a catalogue or
classification of ambiguity as a guide. Analysing ambiguity in this way demonstrates
how ambiguity takes different forms and thus provides lawyers with a catalogue of
various types of ambiguity, which can function as a checklist to assist in analysing
ambiguity to identify all of the meanings of the ambiguous provision. This assists
readers in becoming more proficient at detecting ambiguity, understanding its nature
and framing arguments to resolve it. This is more so the case given that some forms of
ambiguity, for example ambiguity of implication, are not always obvious to an
untrained eye.

On the wide view of its scope ambiguity can be classified into five major types, most
of which contain sub types.9 The five types are lexical ambiguity, relational ambiguity,
ambiguity of implication, ambiguity from competing versions of a rule and ambiguity
from a conflict between rules. Knowing these categories may assist in detecting an
ambiguity, understanding how it has arisen and in formulating arguments to resolve it.
These categories are fully explained in another text, so what now follows is just an
outline of these forms of ambiguity. This is followed by discussion of some cases that
are special because they can be contentious.

Lexical Ambiguity
Lexical or verbal ambiguity10 is ambiguity within words themselves. A word can have
two or more distinct meanings.11 In the phrase ‘offensive behaviour’ does the word
‘offensive’ mean aggressive (as in taking the offensive) or disgusting?12 A word can
also be wide or vague in its terms and hence uncertain in its reach. How high does a

7. Christopher Enright Legal Method Chapter 14 Classifying Meanings

8. Commentary 25.2.
9. See Christopher Enright Legal Method, Chapter 14 Classifying Meanings
10. Repatriation Commission v Vietnam Veteran’s Association (2000) 171 ALR 523, 550 per
Spigelman CJ
11. Ormond Investment v Betts [1922] All ER Rep 709 per Lord Buckmaster
12. R v Smith [1974] 2 NSWLR 586
 Chapter 25 Analysing Ambiguity 385

person have to be to be ‘tall?’ How much is enough? With open-ended expressions

like these the best that a court can usually do is rule, in each particular case as it arises,
on which side of the line the item in question falls.

Relational Ambiguity
Ambiguity can lie in the structure of sentences. This can also be termed ‘syntactic or
grammatical ambiguity’.13 An example is the sentence: ‘The historic house was saved
from destruction by a developer’. It is not clear from this whether the developer saved
the house or was the person who threatened to destroy it. The concluding phrase ‘by a
developer’ could as a matter of syntax attach either to ‘saved’ or to ‘destruction’.

Implication
Ambiguity can arise from implication.14 There is an implication from the text that the
words might be read either more widely or more narrowly than their ordinary
meaning.15 Something is effectively to be added to the text or something is to be taken
away from the text. Implication raises the question of whether the text should or
should not be read as subject to the implication. Justification for making the
implication can be found in the golden rule of interpretation (which is discussed
below).

An example of implication is the rule written on the walls of stations in the London
underground rail system. It says: ‘Dogs must be carried on the escalator’. To analyse
this ambiguity it is necessary to divide those who might travel on the escalator into
two classes. Class 1 consists of those who have a dog with them while Class 2 consists
of those who do not have dog with them. Taken literally, members of both Class 1 and
Class 2 would have to carry a dog to ride the escalator. This means that members of
Class 2 would have to procure a dog by some means or not ride the escalator.

Common sense, however, suggests that the sign has to be read subject to an implied
qualification. Its purpose is to protect dogs from harm. So the implied qualification is
in all likelihood that the sign is meant to apply to people in Class 1 only being those
who have a dog with them.

Competing Versions of a Rule

Ambiguity of competing versions of a rule occurs only with common law. It arises
when, for example, one judge frames a rule with one set of words while another judge
frames the rule with another set of words similar to but not identical with the first.

Conflict Between Rules

In ambiguity based on conflict between rules (in common law or statute), one rule
regulates a situation in one way while a second rule regulates the situation in another
way.
13. Repatriation Commission v Vietnam Veteran’s Association (2000) 171 ALR 523, 550
14. Commentary 25.3.
15. Commentary 25.4.
 386 Chapter 25 Analysing Ambiguity

Special Cases
In this context special cases refer to situations where facts do not precisely fall within
the language of a statute but do so if the provision is stretched. There are at least three
cases. One involves open terms (a form of lexical ambiguity), a second involves partial
satisfaction (a form of lexical ambiguity) while a third involves implied qualification
and implied extension. These raise a question as to the scope of ambiguity, which is
why the text gives them special treatment.

Open Terms
This is a form of lexical ambiguity. Some terms are so vague, wide and subjective that
they effectively confer a discretion on a court. Common examples of such words in
statutes are provisions that authorise some authority to take action that must be
‘reasonable,’ ‘fit and proper,’ or ‘just and equitable’. These terms often raise a
question of degree, for example how just and equitable must it be to satisfy the
requirement? The scope of these terms is chronically wide. Therefore to make these
terms more manageable courts often develop criteria and guidelines to implement
them. As a result such terms, although contained in statute, give rise to a new common
law as courts endeavour to turn their open texture into more definite standards. This is
referred to as statutory common law. Clearly with this type of ambiguity there is no list
of specific meanings but an amorphous spectrum of possibilities.

Partial Satisfaction
Partial satisfaction is a form of lexical ambiguity. Sometimes a fact partially satisfies
the expressed or assumed requirements of a term. There is a literary example in Banjo
Patterson’s poem Clancy of the Overflow, which refers to a letter written by a shearer
with a ‘thumbnail dipped in tar’. Is this a ‘pen’? Or is an inverted wooden packing
case a ‘table’? This ambiguity is sometimes described by reference to words having an
umbra of certainty and a penumbra of uncertainty. Cases that definitely fall within the
term are the umbra. Cases on the fringe such as the thumbnail dipped in tar and the
inverted wooden packing case fall within a penumbra of uncertainty.

Implied Extension and Implied Qualification

Implication takes two forms, implied extension and implied qualification. This can be
explained as a statute’s having a central meaning, an umbra that is uncontentious, and
a possible or arguable wider meaning, a penumbra that is contentious:
(1) Implied Qualification. An implied qualification narrows the scope of a
provision by lessening the area that it covers. It narrows the umbra.
(2) Implied Extension. An implied extension does the reverse. It extends the scope
of a provision by extending the area that it covers. It creates or expands the penumbra.

Scope of Ambiguity
The cases of vague terms, partial satisfaction and implication are important because
they open up the question of the scope of ambiguity. According to a narrow view of
ambiguity these are not proper ambiguities. In the discussion here the text just marks
them for identification. The text discusses the question of the scope of ambiguity
 Chapter 25 Analysing Ambiguity 387

below and in doing so refers back to this description of these three disputed types of
ambiguity.

Scope of Ambiguity
Introduction
While ambiguity makes interpretation necessary,16 somewhat ironically the word
‘ambiguity’ is itself ambiguous or at least has been rendered so by lawyers. In their
wisdom, or more likely their lack of it, they have bestowed on it both a wide and a
narrow meaning. The question therefore is which meaning of ambiguity should courts
adopt. To appreciate fully discussion of this question it may be either helpful or
necessary for the reader to acquaint themselves with the five types of ambiguity –
lexical ambiguity, relational ambiguity, ambiguity of implication, ambiguity from
competing versions of a rule and ambiguity from a conflict between rules – by reading
again the outline of them that is located earlier in this chapter.17

The wide meaning is that for legal purposes ambiguity includes all five types including
their sub-types. The narrower view of ambiguity is to a substantial extent determined
by inference because it is generally implied rather than expressed.18 It has two
components, which have been explained above. One component excludes meanings
that do not fall fully within the text of the law. This encompasses lexical ambiguity
based on partial satisfaction and ambiguity of implication. A second component
excludes lexical ambiguity based on vagueness.

Wide Meaning
The wide meaning is the basis of the analysis of language presented in this chapter.
The wide view is that there is ambiguity when ‘the intention of the legislature is for
whatever reason doubtful’ (which is why it fully includes all of the five types of
ambiguity).19 This is plain logic. A provision is ambiguous for the purposes of legal
interpretation if there is some uncertainty about its meaning, leading to uncertainty as
to whether or not the provision applies to the facts of a case. On one view of the law it
applies, on another view it does not. So, ambiguity exists where by any stretch of the
imagination a provision could reasonably be interpreted and applied one way or
another. In other words, whatever generates the need for interpretation constitutes
ambiguity.

Glanville Williams also takes this view. As he put it, the proper test is to put the
question: Does the provision in the case apply or not apply to particular facts? If on a
reasonable view either a ‘yes’ and a ‘no’ is possible, there is ambiguity. On this line of
reasoning, a court can decide that there is no ambiguity on a point only if it decides in
the context of the case before it that ‘[any] alternative interpretation is impossible on

16. R v L (1994) 122 ALR 464 at 468

17. See Christopher Enright Legal Method Chapter 14 Classifying Meanings.
18. For example, it is reflected in the drafting of s15AB of the Acts Interpretation Act 1901
(Cth) and its counterparts in other jurisdictions.
19. Repatriation Commission v Vietnam Veteran’s Association (2000) 171 ALR 523, 550
 388 Chapter 25 Analysing Ambiguity

the wording’. In other words, the court rules that the meaning that a party seeks to put
on a word or phrase is lexically untenable.20

This wide view of ambiguity would include all of the five types of ambiguity
discussed above. Thus it encompasses lexical ambiguity, relational ambiguity,
ambiguity of implication, ambiguity form competing versions of a rule and ambiguity
from a conflict between rules.

Proposed Exclusion 1: Meaning not Fully Within Text of Law

One strand of the narrow form of ambiguity is based on a proposed meaning of a
provision not falling totally within the text of the ambiguous provision. There are two
specific types:
(1) Ambiguity of Implication. This is by far the most common exclusion,21 and
occurs in texts,22 legislation23 and cases.24 Yet despite this neglect of ambiguity of
implication in formal discussion, there is wide resort to this ambiguity in cases.
(2) Partial Satisfaction. This applies where the meaning falls partly within the
provision and partly outside.

Both of these forms of ambiguity raise the question of how far courts should allow
language to go. Must an item fall squarely with a word or phrase, that is, within the
umbra? Or is it enough that they fall fairly but not squarely or precisely within the
term, that is, within the penumbra?

An extreme literal approach would not countenance ambiguities of this kind. Either the
facts fit the words precisely or they do not fit them at all. This approach is referred to
in the United States as textualism, or in its latter day appearance, the new textualism.
As an illustration the United States Supreme Court decided a case where the relevant
provision provided a higher penalty for purchasing drugs in the case where the
purchaser ‘used’ a gun in the course of the purchase. In this case the purchaser sought
to barter the gun for drugs. Defence counsel argued an implied qualification so that
one ‘used’ a gun for the purpose of the statute when one used it as a firearm. The
Supreme Court, however, refused to impose the implied qualification and so held that
the defendant ‘used’ the gun for the purchase.25

An alternative approach takes into account that language is not an inherently precise
means of communication and that attempts to make it more precise will often cause a
legislature to use more rather than fewer words. Further, where the intention of the
legislature is obvious, rational and just, the legalism entailed in a strict reading of the
scope of words defies common sense and creates unnecessary disruption and disorder.

20. Williams (1981A) p 1129

21. Williams (1981A) , referring to implication as the ‘secondary’ meaning of a provision.
22. An example is Pearce (2001) p 25.
23. Commentary 25.5.
24. Commentary 25.6.
25. Smith v United States 53 113 SCt 2050 (1993)
 Chapter 25 Analysing Ambiguity 389

A problem with this ambiguity is that judicial discussion is not very articulate. Lord
Diplock, for example, refers to judges who ‘invent fancied ambiguities’.26 As another
example, Justice Dawson has drawn a strange distinction between ‘provisions on their
face [which] offer more than one construction’ and ‘determining whether more than
one construction is open’.27

Proposed Exclusion 2: Vagueness

The narrow view may exclude lexical ambiguity to the extent that it arises from words
that are vague or uncertain. For example, s15AB(1)(b)(i) of the Acts Interpretation Act
1901 (Cth) draws a distinction between a provision which is ‘ambiguous’ and one
which is ‘obscure’.

Dickerson also argues for this distinction in the following way: ‘Whereas ‘ambiguity’
in its classical sense refers to equivocation, “vagueness” refers to the degree to which,
independently of equivocation, language is uncertain in its respective application to a
number of particulars. Whereas the uncertainty of ambiguity is central, with an “either-
or” challenge, the uncertainty of vagueness lies in a marginal question of degree’.28

Resolution
This chapter adopts the wider view. The best reason for taking the wider view is that
justice is generally better served by allowing courts to take the wider view. While it
may involve some damage to the proper use of language it is avoids unnecessary
complications with only minor cost.

There is some general judicial support for the wider view.29 The golden rule of
interpretation also provides some support for the wider view in relation to ambiguity
of implication and ambiguity based on partial satisfaction. It applies where the literal
sense of a statute would ‘lead to some absurdity, or some repugnance or inconsistency’
with the rest of the statute.30 These, it should be noted are fairly extreme cases.31 When
it operates, the golden rule allows a court to depart from the ordinary meaning of the
language of a statute by bending the meaning of a provision by resorting to
implication32 or by allowing a meaning that only partially fits within the words of the
statute.

Yet the spirit with which courts go beyond the literal scope of the words should be the
same in each case. A court cannot pluck something out of thin air and by this means
break free of the constraints of the words. It can, however, bend or stretch the words
but only up to a point. In the result it gives the words a permissible but ‘strained’33 or
26. Duport Steels v Sirs [1980] 1 WLR 142, 157, [1980] 1 All ER 529
27. Mills v Meeking (1990) 91 ALR 16, 30-31. Commentary 25.7.
28. Dickerson (1964) p 10
29. Commentary 25.8.
30. Grey v Pearson (1857) 6 HLC 61, 106 per Lord Wensleydale
31. Commentary 25.9.
32. Commentary 25.10.
33. Bermingham v Corrective Services Commission (1988) 15 NSWLR 292, 302
 390 Chapter 25 Analysing Ambiguity

‘ambulatory’34 construction, but this construction should be neither ‘unreasonable nor

unnatural’.35

Prevalence of Ambiguity
How prevalent is ambiguity? Postmodernism, which is often invoked to explain legal
interpretation (sometimes referred to in the context of interpretation as
conventionalism),36 propounds the view that language is extremely open to ambiguity.
Commonly its exponents refer to the ‘indeterminacy’ of legal language.37 Affiliated
with this is the notion that each of us is likely to take our own interpretation. As the
Latin proverb puts it, quot homines tot sententiae,38 that is, there are as many opinions
as there are people. Thus, Berns writes that ‘interpretation is a profoundly creative act
[because] to read a text is also to create that text anew,’39 and Wald asserts that
‘language is inherently indeterminate and will always depend upon both the writer and
the reader’s context to give it meaning’.40 In a similar vein Hart, a positivist we should
add, says that language is ‘open textured’.41

One can, however, test this view by going to the law reports and looking at any case
interpreting a statute or a common law rule. In such cases there is no suggestion or
concern that the ambiguous word or phrase has a multitude of meanings. Instead it has
a finite number of meanings, usually just a handful, or in the case of an open
expression such as ‘just’ or ‘fair’ an identifiable spectrum of meanings. What troubles
the court is how to decide between these meanings. Given that this is the approach
frequently adopted by courts, postmodernists’ views on the nature of language seem
both extreme and irrelevant.42

A more conventional view is that ambiguity is common enough but not overwhelming.
Professor Julius Stone took this approach. Ambiguity provides choice but not open
slather. It merely gives a court ‘leeways of choice’43 on some occasions, even if in
some cases this ‘creative choice’ is ‘secret and even unconscious’.44

The view taken here is similar to Julius Stone’s. Language is potentially ambiguous
but is not always so in all contexts. Except in the case of a wide and open term with an
identified spectrum of possibilities, there is a finite number of meanings. One side to
the case will present some meanings, the other side may present some and the court

34. R v Young (1999) 46 NSWLR 681, 687-688

35. IW v City of Perth (1997) 146 ALR 696, 702. Commentary 25.11.
36. Schanck (1990) pp 833-843
37. See Greenwalt (1990); Endicott (1996) and Drahos and Parker (1991).
38. Publius Terentius Afer (185 BC – 159 BC), a Roman comic dramatist (aka Terence)
39. Berns (1993) p 63
40. Wald (1990) p 302
41. Hart (1961) p 125
42. Commentary 25.12.
43. Stone (1964) p 276, and see Queensland v Commonwealth (1977) 139 CLR 585, 603 per
Stephen J, 606 per Mason J.
44. Stone (1964) p 241. See also Krygier (1986).
 Chapter 25 Analysing Ambiguity 391

may present possible meanings that the parties have not conceived. Even if there are
other possible meanings, if they are not raised in court by one or other party or the
judge they do not count in the particular case (although they may be raised in a later
case). This finite list of the possible meanings of an ambiguous provision is the range
of options or possibilities before the court. The court can only choose from among this
list to come up with the correct answer.

Relevance of Ambiguity
Introduction
When a lawyer or court is confronted with a question of interpretation logically one of
the first things that they need to do is to identify the meanings of the ambiguous
provision and the effects that each will cause. As described above, this is Step 1 in the
model for interpreting law. For reasons given below, this is of major importance.
Unless one can analyse language to identify the various meanings of an ambiguous
provision it is not possible to interpret law either as a lawyer or a judge. Trying to
interpret law without identifying ambiguity in biblical terms is akin to building bricks
without straw.

While it might seem intuitively right to identify the meanings of the ambiguous
provision and their effects it will enhance the interpretive skills of readers if the
reasons for this are articulated. There are in fact three major reasons for identifying the
options. It defines the problem, it identifies the solution to the problem, and it enables
reasons to target the particular meaning to which they must be addressed.

Defining the Problem

Identifying all of the meanings of the ambiguous provision defines the problem. When
interpreting law the problem arises from a provision of a statute being ambiguous. Step
1 defines the problem by identifying each of the various meanings of this ambiguous
provision. The existence of these meanings generates the ambiguity. By identifying
these meanings the court knows precisely what the problem is. By identifying the
effects that each meaning will cause the court knows the significance of the problem.
When the court chooses one of these meanings as legally correct it is implementing the
effect that that meaning will cause. At the same time it is rejecting the effects that the
other meanings would cause if chosen as correct.

Thus, identifying the meanings and their effects makes clear what choice the
ambiguity has imposed on the court. It must decide which of several competing
meanings of the ambiguous provision is legally correct and by this means introduce
into the world the effect that this particular meaning will cause.

Identifying the Solution

If the lawyer or the court correctly identifies all possible meanings of the ambiguous
provision it also identifies the solution to the problem – it must be one or more of
those meanings. This is how the model seeks to ensure that the task is done
comprehensively so that nothing of relevance is overlooked. In consequence, one of
 392 Chapter 25 Analysing Ambiguity

the options must constitute the best available solution to the problem.45 Conversely, if
not all options have been identified there is always the possibility that the best
outcome consists of an option that has not been identified – in consequence it will not
be appraised by the court in Step 2 and therefore cannot be indorsed in Step 3 as the
legally correct meaning of the provision.

Assisting the Reasoning Process

Identifying all of the meanings of the ambiguous provision is necessary for the
reasoning process, which takes place in Step 2 of the model. This reasoning consists of
arguments addressed for or against options. Any reason or argument addressed to the
court makes sense only if it is directed to a meaning of the ambiguous provision. It
seeks to persuade the court to accept or reject that meaning as the legally correct
meaning of the ambiguous provision. Obviously, it is not possible to formulate an
argument for or against a meaning until the meanings have been precisely identified.

2. Limits to Interpretation
Ambiguity provides an interpreting court with a choice. This choice, however, is not
unlimited. Conveniently the limits can be stated in two phases. There are general
limits, which are imposed by the fact that any interpretation made by a court must fit
broadly or reasonably within the limits of the language used in the ambiguous rule.
Within this broad limit there are specific limits imposed by the various types of
ambiguity. Each ambiguity confers a choice, while at the same time imposing limits on
that choice.

This analysis of language is directly relevant to interpreting law. It is also relevant to

any consideration of the contentious question of judicial activism. Obviously the limits
of interpretation set outer limits to the creative role of the courts when interpreting law
(but not when making common law). The question then is whether there are any
ascertainable inner limits that define the proper creative role of a court.

General Limits: Language

[U]nless the statute plainly hands courts the power to create and revise a form of common law,
the domain of the statute should be restricted to cases anticipated by its framers and expressly
resolved in the legislative process.46
Introduction
The proposition that language sets limits on interpretation flows from two other
propositions. Here are all three of these propositions:
(1) Ambiguity as the Problem. Ambiguity is the source of a problem because it
gives rise to the need for interpretation. An ambiguous word or phrase has two or more
meanings and a court must decide which is legally correct.
(2) Ambiguity as the Answer. Ambiguity, however, is also the source of the
answer. The answer to the question of interpretation lies within these meanings.

45. Commentary 25.13.

46. Easterbrook (1983) p 544
 Chapter 25 Analysing Ambiguity 393

(3) Limits of Language. Logically, a court can choose its answer to the question of
interpretation only from the meanings that the ambiguous word or phrase properly
yields up. This is how language sets limits on the function of interpretation. If the
language of the ambiguous law does not encompass a meaning of a provision then it is
not available as the legally correct interpretation.

This proposition that language sets limits on the task of interpretation is actually
framed in the literal rule of interpretation, which however, is one of the most
misstated, and in consequence, misunderstood and misapplied rules in the entire
common law. Stated simply in its conventional form, the rule says that words should
be given their literal meaning.47 As commonly understood, the literal rule is considered
to guide courts in the process of interpretation by pressing for the literal meaning of a
word or phrase as the correct legal meaning.48 By contrast, the view propounded here
is that this is not the proper sense of the literal rule. However, badly expressed as it is,
the only way to make sense of the literal rule is to see it as imposing a stricture on
interpretation. This stricture consists of a general requirement that any meaning that a
court chooses as legally correct must fall fairly (even if not totally squarely) within the
ambit of the ambiguous words that the court is interpreting.

Conventional Statement of the Rule

In its common formulation, the literal rule says that words should be given their literal
meaning.49 Other expressions besides literal50 are sometimes used such as natural,
ordinary and grammatical,51 but the effect is the same.

While the literal rule specifically says words should be given their literal meaning
commonly it is taken to imply a further proposition, which has two connected parts:
(1) Part 1. Words possess just one literal meaning.
(2) Part 2. One can discover this literal meaning by objectively means.52 As the
Latin maxim puts it, lucet ipsa per se – the meaning comes shining through.53 Words,
as it were, mean what they say and say what they mean.54

Understood in this way, the literal rule has major implications for the practice of
interpretation. To interpret a statute a court just reads the words of a statute in their
plain, natural or literal sense, sees what they mean and in this way ascertain the answer
to any question of interpretation.55 In other words, someone does not have to be a
lawyer to interpret law – they only need to know how to read.

47. Avel Pty Ltd v Attorney General (1987) 11 NSWLR 126, 127 per Kirby P
48. Commentary 25.14.
49. Avel Pty Ltd v Attorney General (1987) 11 NSWLR 126, 127 per Kirby P
50. Project Blue Sky v ABA (1998) 153 ALR 490
51. Vacher v London Society of Compositors [1913] AC 107, 149. Commentary 25.15.
52. Kirby (2003) in Sheard (2003) p 45
53. Amalgamated Society of Engineers v Adelaide Steamship Co (Engineers’ Case) (1920) 28
CLR 129, 149
54. Repatriation Commission v Kohn (1989) 87 ALR 111, 523 per Hill J. Commentary 25.16.
55. Avel Pty Ltd v Attorney General (1987) 11 NSWLR 126, 127. Commentary 25.17.
 394 Chapter 25 Analysing Ambiguity

Problem
There is a major problem with the literal rule when it is framed in this way because the
basic proposition that it is assumed to imply is not merely questionable but downright
wrong. It is taken to imply that each word has one, single, identifiable meaning which
can be ascertained in an uncontentious way.56 This is not the case for at least two
reasons. First, many words have more than one meaning as flipping through the pages
of a dictionary will reveal.57 This is called lexical ambiguity. Second, there are, as
already discussed, other forms of ambiguity besides lexical ambiguity.

So, for judges to look to just the words of a statute it is not really a means of resolving
ambiguity. Resort to the words of a statute via the literal rule is not a solution to the
problem, but is at best a restatement of it. Given this, as it is conventionally formulated
the literal rule is next to useless. This is why it is necessary to propose a reformulation
of the rule to express it in a coherent form that makes a genuine contribution to the
task of statutory interpretation.

Restatement of the Rule

The Constitution is not a living organism, it is a legal document. It says some things and doesn't
say other things.58
Many words have not just one meaning, but a range of meanings. This easily makes
language loose, flexible and ambiguous. So, the only rational point to the literal rule is
not to assert that there is only one literal and identifiable meaning which answers any
issue of interpretation. Rather, it accepts that words can have several meanings but at
the same time asserts forcefully that the various meanings of a word depict its limits.
These meanings may themselves be fuzzy and require interpretation. Nevertheless, the
fact that a word has several meanings does not indicate that there are no limits to its
meaning. Rather, those meanings together set the outer limits for the scope of the
word, even if these meanings need to be clarified by interpretation. That is, for any
particular text there are limits ‘beyond which the words will not stretch’.59 In this way
the rule both assumes and requires that the intention of the legislature is found only in
‘the language in the statute as whole’.60 Words used in the text of a statute constitute ‘a
constraint on the range of interpretive options available’.61 This is why a court must
not do ‘violence’ to ‘the language of an Act so as to bring within its terms articles
which are not expressly included’.62

Thus it is a fundamental rule that a court may give a meaning to words only if these
words are ‘reasonably open to such a construction’; in this way the language of the
particular law sets limits to interpretation, and does so for the good and simple reason

56. Commentary 25.18.

57. Commentary 25.19.
58. Justice Antonin Scalia (2006)
59. FCT v Trustees of Lisa Marie Walsh (1983) 48 ALR 253 at 278. Commentary 25.20.
60. Amalgamated Society of Engineers v Adelaide Steamship Co (Engineers’ Case) (1920) 28
CLR 129, 161-162, per Higgins J. Commentary 25.21.
61. Eskridge (1990) p 626
62. Corkery v Carpenter [1950] 1 KB 102, 106
 Chapter 25 Analysing Ambiguity 395

that construction of an Act ‘must be text based’.63 As Lord Steyn succinctly put it, ‘the
primacy of the text is the first rule of interpretation’64 so that the text becomes ‘the
formal focus of interpretation’.65

Putting all this in its plainest form, any meaning that a court adopts as the legally
correct interpretation of a provision must fall reasonably within the scope of the words
in the statute. A court should not choose a meaning of a provision as legally correct if
this meaning is not found within the ordinary meaning of the words, grammar and
syntax used by the statute.66 Consequently, those who criticise the interpretation might
say that it is not the best interpretation, but they should never be able to say that the
interpretation chosen by the court was not open to being so chosen. If they can
truthfully assert that this interpretation was not available, it is deeply flawed.

Rationale: Supremacy of the Legislature

Opposition to violence committed on statutory language by courts is based on the
supremacy of the legislature. It is for the legislature and not the courts to change the
language of a statute.67 This is why it is fundamental that all of the legitimately
possible meanings of an ambiguous provision ‘must be found in the words of [the
legislature]’.68 That is, the ‘mandated point of departure’ or commencement when a
court starts out on the road to interpretation ‘must be the text itself’.69

Notwithstanding the supremacy of the legislature, there is an argument that a court can
or should depart from the literal meaning of the words of a statute when doing so
implements the obvious but unstated intention of the legislature. But compelling as
this case is, a court should not do violence to language, even if its aim is to achieve an
established or assumed intent of a legislature.70 A court cannot add words to a statute
or rewrite it when, through oversight or inadvertence, the clear intention of the
legislation has not been translated into the text of the law.71 Interpretation is deciding
which of two or more meanings is legally correct. It is not ‘a warrant for redrafting
legislation nearer to an assumed desire’ of the legislature.72 As Easterbrook neatly puts
it, in these cases a court should say to a litigant: ‘Too bad, but legislative intentions are
not legal rules’.73 One sticks to the letter of the law, and must not depart from it in
pursuit of the supposed spirit or purpose of the legislation.74 Putting it simply, law is
the language and the language is the law.
63. R v Young (1999) 46NSWLR 681, 687-688
64. Steyn (2002) p 5
65. Eskridge (1990) p 626
66. Corkery v Carpenter [1950] 1 KB 102, 104, Federal Commissioner of Taxation v Trustees of
Lisa Marie Walsh (1983) 48 ALR 253, 278
67. R v The Judge of the City of London Court [1892] 1 QB 273, 301-302 per Lopes LJ
68. Trevisan v FCT (1991) 101 ALR 26, 31
69. Steyn (2002) p 5
70. Pearce (2001) pp 20-22
71. Brennan v Comcare (1994) 50 FCR 555; 122 ALR 615
72. Trevisan v FCT (1991) 101 ALR 26, 31
73. Easterbrook (1983) p 534
74. Indeed the English word ‘literal’ is derived from the Latin word littera, a letter.
 396 Chapter 25 Analysing Ambiguity

Another possibly compelling reason for departing from the literal meaning of the
words might be that it furnishes a means of avoiding a result that is undesirable, for
example because it is ‘inconvenient or impolitic or improbable’.75 Again, though, there
is the standard reply. It is for the legislature not the court to judge how wise, or just or
useful a provision is.76 A court must still not do violence to the language of a statute;
instead it must give effect to the words of a statute, no matter how ill conceived it may
be.77 In any event, a policy is not ill conceived because a judge disagrees with it.78

Moreover, while the literal rule generally confines a court to meanings that properly
fall within the language of the statute, it is worth emphasising that within these
confines there are presumptions that strain towards doing justice. While a court should
seek to implement the intention of the legislature, the legislature is assumed to intend
to legislate in a way that is fair, just, reasonable and respecting of human rights.
Consequently a court will endeavour to interpret a statute to achieve these outcomes
unless the language of the provision clearly indicates a contrary intention or somehow
precludes the statute from being interpreted in this way.

Exceptions
In prior discussion in this chapter we considered two special types of ambiguity –
ambiguity based on partial satisfaction and ambiguity based on implication. In both of
these cases courts accept as proper meanings the meaning derived from partial
satisfaction and implication where the meaning fitted fairly but not squarely within the
words of the provision. These operate as exceptions or qualifications to the literal rule.

Comment
Relevant to our analysis is that many lawyers are not fully conscious of implied
qualification and implied extension to a statute. Consequently, they fail to know and
appreciate the only meaning of the literal rule that makes sense. This failure to
recognise qualification or extension of a statute explicitly occurs in texts, cases79 and
legislation,80 despite widespread resort implication in cases. Confusion ensues, and
this confusion generates a substantial amount of muddled comment about the nature of
statutory interpretation.

Specific Limits: Ambiguity

As has just been discussed, the words of the common law or statutory rule that is
ambiguous set outer limits to the scope of the statute. These limits are stated in the
fundamental notion that the language is the law and the law is the language, which

75. Amalgamated Society of Engineers v Adelaide Steamship Co (Engineers’ Case) (1920) 28

CLR 129, 161-162, per Higgins J. Commentary 25.22.
76. Phosphate Co-op v Environment Protection Authority (1977) 18 ALR 210
77. Davies v Deverell (1992) 1 Tas R 214, Shire of Arapiles v Board of Land and Works (1904) 1
CLR 679
78. McCawley v The King (1918) 26 CLR 9, 24 ALR 413
79. Commentary 25.23.
80. Commentary 25.24.
 Chapter 25 Analysing Ambiguity 397

leads to the proposition that a court may give a meaning to words only if these words
are ‘reasonably open to such a construction’.81

Within this broad limit there are further limits imposed by the various types of
ambiguity. According to the classification deployed in this book, there are five types
of ambiguity:
# lexical ambiguity
# relational ambiguity
# ambiguity of implication
# ambiguity from competing versions of a rule
# ambiguity from a conflict between rules.
Each form of ambiguity confers a choice on a court. At the same time they define and
impose limits on that choice.

These forms of ambiguity are outlined above. They are discussed in detail in another
text, which also explains both the leeway that they furnish for interpretation and the
limits that they impose.82

Commentary
Commentary 25.1 Footnote 2
For consideration of ambiguity by writers see Evans (1989), Drahos and Parker
(1991), Maher (1984), Endicott (1996), Slattery (1996), Bennion (1980), Bennion
(1981A) and Bennion (1981B).

Commentary 25.2 Footnote 8

Empson ‘Seven Types of Ambiguity’ (1930) is a classic text. It deals with ambiguity
from a literary perspective, showing how ambiguity of feeling creates imagery and
affects our senses. It is not, however, of specific or direct relevance to legal
interpretation.

Commentary 25.3 Footnote 14

Some examples of ambiguity of implication are found in KP Welding Construction Ltd
v Herbert (1995) 102 NTR 20, 40-41 and Peninsula Group v Registrar-General of the
Northern Territory (1996) 136 FLR 8, 12.

Commentary 25.4 Footnote 15

There is a special case for ambiguity of implication that occurs where there is
ambiguity of competing rules. This is a special case because it is created by the
legislature in making two inconsistent rules. The court has to resolve the conflict by
doing two things. It reads down (that is reduces) the scope of one or both rules to allow
both to operate. In doing so it imposes an implied qualification on the scope of the rule
or rules that it reads down.

81. R v Young (1999) 46NSWLR 681, 687-688. Commentary 25.25.

82. Christopher Enright Legal Method Chapter 14 Classifying Ambiguity
 398 Chapter 25 Analysing Ambiguity

Commentary 25.5 Footnote 23

Section 15AB(1)(b) (i) and (ii) of the Acts Interpretation Act 1901 (Cth) reflect this
distinction between the wide and narrow meaning of ambiguity. Section 15AB(1)(b)(i)
reflects the wider meaning as it refers to a provision that is ‘ambiguous’.
Section15AB(1)(b)(ii) reflects a narrower meaning as it refers to a provision whose
‘ordinary meaning’ leads to a result that is ‘manifestly absurd’ or ‘unreasonable’.

Commentary 25.6 Footnote 24

In relation to ambiguity of implication note in Mills v Meeking (1990) 91 ALR 16, 30-
31 Dawson J drew a distinction between ‘provisions on their face [which] offer more
than one construction’ and ‘determining whether more than one construction is open’
(which is a roundabout way of referring to ambiguity of implication). A similar
attitude was expressed in two cases cited by Dawson J, Miller v Commonwealth (1904)
1 CLR 668, 674 and Wacal Development Realty Development Pty Ltd (1978) 20 ALR
621, 630. For other judicial use of this approach see CIC Insurance Ltd v Bankstown
Football Club Ltd 141 ALR 618, 634-635 per Brennan CJ, Dawson, Toohey and
Gummow JJ; Isherwood v Butler Pollinow (1986) 6 NSWLR 363, 368 and Attorney
General v Prince Ernest Augustus of Hanover [1957] AC 436, 461.

Commentary 25.7 Footnote 27

Mills v Meeking (1990) 91 ALR 16, 30-31 makes this strange distinction between
‘provisions on their face [which] offer more than one construction’ and ‘determining
whether more than one construction is open’. This distinction appears in two cases
cited by Dawson J, Miller v Commonwealth (1904) 1 CLR 668, 674 and Wacal
Development Realty Development Pty Ltd (1978) 20 ALR 621, 630.

Commentary 25.8 Footnote 29

There is some general judicial support for the wider view in Repatriation Commission
v Vietnam Veteran’s Association (2000) 171 ALR 523, 550 per Spigelman CJ and in
Bowtell v Goldsborough Mort & Co (1996) 3 CLR 444 at 456-457, per O’Connor J.
Sir Samuel Griffith neatly put support for implication as a form of ambiguity in Nicol
v Chant (1909) 7 CLR 569, 581, in saying: ‘There is no word the primary meaning of
which may not be modified by the context’.

Commentary 25.9 Footnote 31

As the text says, the golden rule applies where the literal sense of a statute would ‘lead
to some absurdity, or some repugnance or inconsistency’ with the rest of the statute.
This means that the golden rule should not be deployed where the meaning is clear but
the result is not as just as the court would have it. Here, the operative maxim is stant in
claris (literally, they rest in clarity), that is, when the meaning is clear look no further.
The maxim applies the principle of Occam’s razor to statutory interpretation.

Commentary 25.10 Footnote 32

There are rules for determining when an implied qualification or extension is
permissible – see Christopher Enright Legal Method Chapter 14 Classifying Meanings.
 Chapter 25 Analysing Ambiguity 399

Commentary 25.11 Footnote 35

In Luke v Inland Revenue Commissioners [1963] AC 557, 577 Lord Reid expressed a
similar sentiment to IW v City of Perth (1997) 146 ALR 696, 702. There he intimated
that the outer limit of interpretation was when the words in the legislation were
‘absolutely incapable’ of a proposed meaning. Up to this point, the meaning was
legitimate (although whether it was the legally correct meaning of course is a separate
question).

Commentary 25.12 Footnote 42

Greenwalt (1990) argues that ‘law often yields determinate answers’ and in so doing
adopts ‘a fairly strict interpretation’ of what constitutes a determinate answer (p 2).

Commentary 25.13 Footnote 45

When interpreting law, a comprehensive statement of all of the meanings of the
ambiguous provision will be an organising framework on the basis that courts do not
do violence to language. They must not give the words a meaning that they cannot
reasonably bear. This in fact, is the proper meaning of the literal rule despite much
common misunderstanding to the contrary.

Commentary 25.14 Footnote 48

For conventional accounts of the literal rule see Bennion (1980A), Williams (1981A)
and Williams (1981B). For comparison between literal and purposive interpretation
see Edit (1981). In the application of the literal rule to the Commonwealth Constitution
see Coper and Williams (1997B), Craven (1992A) and Craven (1992B).

Commentary 25.15 Footnote 51

For cases discussing the literal rule see Australasian Temperance and General Mutual
Life Assurance Society v Howe (1922) 21 CLR 290, 302 and Project Blue Sky v ABA
(1998) 153 ALR 490.

Commentary 25.16 Footnote 54

There are three connected propositions here:
(1) The comment that words mean what they say and say what they mean was
made by Justice Hill in Repatriation Commission v Kohn (1989) 87 ALR 111, 523.
(2) This echoes what the court said in Connecticut National Bank v Germain 112 S
Ct 1146, 1149 (1992): ‘[I]n interpreting a statute a court should always turn to one
cardinal canon before all others – courts must presume that a legislature says in a
statute what it means and means in a statute what it says there’.
(3) These statements reflect the statement already made on several occasions in this
chapter: ‘The language is the law and the law is the language’.

Commentary 25.17 Footnote 55

This discussion of the commonly alleged but deeply flawed meaning of the literal rule
brings to mind a comment by a non-lawyer critic of a distinguished judge who
 400 Chapter 25 Analysing Ambiguity

espoused a legalistic approach to interpretation. The critic commented that the judge
was regarded as great only because he could read a dictionary.

Commentary 25.18 Footnote 56

The argument that the words of a statute can yield an objective meaning, it is worth
pointing out, is in sharp contrast to looking at the policy of the statute. In adopting this
‘objective meaning’ approach courts eschew the actual intention of the makers and
look to the words themselves. The problem with this approach is that proceeding in
this way will rarely solve a question of interpretation.

Commentary 25.19 Footnote 57

There is a good illustration of how words have multiple meanings in London and
North Eastern Railway Co v Berriman [1946] AC 278. There, all five members of the
House of Lords said that the meaning of an Act was plain, but the plain meaning was
one thing for two of the law lords and another thing for the other three. See also
Maunsell v Olins [1975] AC 373, 385-386 per Lord Wilberforce.

Commentary 25.20 Footnote 59

For some famous cases where the literal rule was invoked see Re Rowlands [1963] 1
Ch 1, Magor and St Mellons v Newport Corporation [1951] All ER 839, and Whiteley
v Chappell (1868-1869) 4 LQRB 147.

Commentary 25.21 Footnote 60

The text asserts the proposition that the true meaning of a provision of a statute must
be properly found in or justified by the words of the statute. A similar view was
expressed in Caminetti v US 242 US 470 (1917) which asserted that ‘the meaning of a
statute must, in the first instance, be sought in the language in which the act is framed’.

Commentary 25.22 Footnote 75

For an extreme implementation of the literal rule see Higgins v O’Dea [1962] WAR
140.

Commentary 25.23 Footnote 79

An illustration is Steyn (2002) pp 3-4, where His Lordship said: ‘The purpose of
interpretation is sometimes mistakenly thought to be a search for the meaning of
words. This in turn leads to the assumption that one must identify an ambiguity as a
pre-condition to taking into account evidence of the setting of a legal text’. See also
Pearce (2001) p 25 and Steyn (2001).

Commentary 25.24 Footnote 80

Dawson J, for example, in Mills v Meeking (1990) 91 ALR 16, 30-31 drew a
distinction between ‘provisions on their face [which] offer more than one construction’
and ‘determining whether more than one construction is open’ (which is a roundabout
way of referring to ambiguity of implication). A similar attitude was expressed in two
cases cited by Dawson J, Miller v Commonwealth (1904) 1 CLR 668, 674 and Wacal
 Chapter 25 Analysing Ambiguity 401

Development Realty Development Pty Ltd (1978) 20 ALR 621, 630. See also CIC
Insurance Ltd v Bankstown Football Club Ltd 141 ALR 618, 634-635 per Brennan CJ,
Dawson, Toohey and Gummow JJ; Isherwood v Butler Pollinow (1986) 6 NSWLR
363, 368; Attorney General v Prince Ernest Augustus of Hanover [1957] AC 436, 461,
Newcastle City Council v GIO (1997) 149 ALR 623, 639-643, citing Lord Diplock in
Jones v Wrotham Park Estates [1980] AC 74, 105 and Sarasawati v R (1991) 100
ALR 193, 207, per McHugh J.

Commentary 25.25 Footnote 81

Section 15AB(1)(b) (i) and (ii) of the Acts Interpretation Act 1901 (Cth) reflect this
distinction. Section 15AB(1)(b)(i) refers to a provision that is ‘ambiguous’ while
s15AB(1)(b)(ii) refers to a provision whose ‘ordinary meaning’ leads to a result that is
‘manifestly absurd’ or ‘unreasonable’. Thus s15AB(1)(b)(i) invokes the narrow view
of ambiguity, while s15AB(1)(b)(ii) refers to the circumstances when resort to
ambiguity of implication is permissible.