What about the non-legal facts: Revising Allen and Pardo’s analytical distinction between law and fact

Introduction

By argumentation in the Myth, Allen and Pardo question the possibility of an intentional distinction between law and fact; whether it is ontological, epistemological or analytical. Instead they argue that the distinction is functional, that is, extensional.

The extension of a concept is the set of all instances to which the concept can be truly applied (Putnam, 1975: 216 f). The distinction between the extension and intension is described by Hilary Putnam:

[…] consider the compound terms ‘creature with a heart’ and ‘creature with a kidney.’ Assuming that every creature with a heart possesses a kidney and vice versa, the extension of these two terms is exactly the same. But they obviously differ in meaning. Supposing that there is a sense of ‘meaning’ in which meaning = extension, there must be another sense of ‘meaning’ in which the meaning of a term is not its extension but something else, say the ‘concept’ associated with the term. Let us call this ‘something else’ the intension of the term. The concept of a creature with a heart is clearly a different concept from the concept of a creature with a kidney. Thus the two terms have different intention. (Putnam, 1975: 216 f)

First thing first: Law is fact

Allen and Pardo argue that there is no ontological distinction between law and fact as there is no difference in their capacity of being true or false. ³ First, Allen and Pardo discuss the possibility that there is a distinction between law and fact that coincides with the meaning of ‘law’ and ‘fact’ in everyday language. An important characteristic of this distinction is understood to be that propositions of facts are true or false whereas propositions of law lack the capacity of being true or false. Allen and Pardo argue against this understanding of law and fact as they claim that law is fact, too. Propositions of law are claimed to have the capacity of being true or false, just like propositions about fact.

To say that legal issues are factual issues, in this sense, then, is merely to say that there is such a thing in the world referred to as ‘the law,’ that the law ‘exists,’ ‘is,’ ‘is one way rather than another,’ or that a given answer to a legal question ‘is true’ or is the case. The point is fundamental, yet often overlooked: to the extent one can say that ‘the law is Y’ or ‘rule Y applies’ one can also say ‘it is a fact that the law is Y’ or ‘it is a fact that rule Y applies’. In short, the answers to legal questions are propositional statements with truth value and are therefore, like other propositions with truth value, factual. ⁴

Allen and Pardo’s claim that law is fact is contrasted against what they identified as the ‘court position’; that what is law is not factual in nature. ⁶ However, Allen and Pardo criticise the court for not motivating this view:

The Court argued that punitive damages do not, like compensatory damages, involve questions of historical or predictive facts. The court did not explain why ‘facts tried by a jury’ in the Seventh Amendment refers only to historical or predictive facts. (Allen and Pardo, 2003a: 1774)

From the claim that law is fact, law creation in courts stands out as problematic, according to Allen and Pardo. We are introduced to the problem concerning law creation:

Some claim that plainly the law is not ‘factual’ as it emanates from norms of various kinds and often does not preexist the decision: the ‘law’ only comes into being as a consequence of the decision itself. (Allen and Pardo, 2003a: 1790)

What the distinction is not: The argument against intentional distinctions

As an alternative ontological distinction between law and fact, Allen and Pardo discuss and repudiate Posner’s suggestion that the law-fact distinction is one between social and natural facts. ⁸ They claim that the distinction cannot be the basis for the law/fact distinction as some human-made entities are recognised as fact, examples of this is consideration in contract and substantial performance (Allen and Pardo, 2003a: 1794). Hence, the distinction between social and natural fact is distinct from the legal distinction between law and fact as they have different extensions. There are social facts that are not law.

Introducing legal and non-legal facts

Allen and Pardo’s functional distinction

In the article ‘The myth of the fact-law distinction’, Allen and Pardo discuss several possibilities to distinguish between fact and law. They state that the distinction has some important functions in law;

These three variables are (1) standard conventions concerning the meaning of ‘law’ and ‘fact,’ (2) the judge-jury relationship, and (3) the distinction between mattes of general import and highly specific and localized phenomena. (Allen and Pardo, 2003a: 1770)

In the introduction of the Myth, Allen and Pardo summarise their argument:

We explain the sense in which legal and factual issues are both ontologically and epistemologically similar. Finally, we discuss the possibility of a coherent, analytical distinction, and conclude that no useful analytical distinction exists; the decision to label an issue ‘law’ or ‘fact’ is a functional one based on who should decide it under what standard, and is not based on the nature of the issue. (Allen and Pardo, 2003a: 1771)

When Allen and Pardo state that the law/fact distinction is functional, it means that the question is allocative—that the categorisation is based on who is in the best position to answer the question.

[…] The only distinction is a functional one: namely, that the concepts ‘law’ and ‘fact’ refer to which body does or should decide an issue or under what standard the issue is or should be reviewed (Allen and Pardo, 2003a: 1790)

Matters concerning negligence are labelled as facts in US proceedings, whereas interpretation of contracts is labelled as law. This is considered by Allen and Pardo when discussing the parts of the functional distinction. The question of better knowledge and desire for uniformity and predictability concretises the question of who is in a better position to answer a question:

The extent to which pragmatic considerations determine the allocative question is plain in these areas [commercial law, contracts and negligence, my comment]. Perhaps judges are better fact-finders in commercial practices—because of complexity, their knowledge of the Code or commercial practices, or the desire for uniformity and predictability—but this does not make ‘legal’ issues out of factual issues unless the term simply refers to those issues better decided by one decision maker than another (which, actually, is our view) (Allen and Pardo, 2003a: 1783)

The circularity of intentional functional distinctions

A circular argument is logically valid, but it provides no additional justification for the conclusion. If conclusion S is included in the premises of the argument, the argument is logically circular and provides no reason to believe S. If conclusion S is derived from premise P, and premise P is itself justified by presupposing S, the argument is epistemically circular. An epistemically circular argument ‘cannot be used to rationally produce conviction’ (Alston, 1986: 8 ff, 15). Thus, in order to conclude whether a circular argument is viciously circular, one should ‘take account of what the argument presumes to show by its use of the specific sentences’ (Psillos, 1999: 82).

Circularity may also render definitions useless. A circular definition cannot inform a person of any meaning of the term of which the person was not already aware. If concept A is defined by reference to concept B, but concept B, in turn, is defined by reference to concept A, then this definition is circular (Murphy, 2002: 17).

It has been argued specifically that reference to better knowledge is circular. The functional definition has been claimed to be circular in general as well. The questions of whom is the better suited to answer the question have been explicitly discussed by Joseph W Smith. He argues that questions of law cannot solely be described as the questions which the court is in a better position to decide. He points out that this definition does not answer when the court is in a better position, and claims that this must be when the court has greater interpretative competence, which has to be when the question is one of law. Therefore, it is impossible to avoid the definition of questions of law by referring only to letting the one ‘best suited’ answer the question (Smith, 2009: 74).

The objection that the functional definition is circular is put forward already by Thayer in his classical article on fact and law from 1890. He writes:

The discrimination of law and fact, in its relation to jury trials, is often identified by practitioners, judges, and law-writers, with the question of what matter is for the court and what for the jury. […] But if we ask the question what sort of thing it is that is for the court and what for the jury, we do not get on, for we are told that matters of law are for the court, and that matters of fact for the jury,—ad quæstionem juris respondent judices, ad quæstionem facti respondent juratores. We do not, then, escape the necessity of trying to determine what a matter of fact is and what a matter of law is. (Thayer, 1890: 147)

When the functional distinction is used as an an intentional definition of the law/fact distinction it leads to circularity. The court has to classify issues as of law or of fact in exactly those cases when a court has to decide whether the right to jury-trial applies to a certain question, or whether a certain question could be tried by the court of appeal. When the court has to decide if a certain question is one of fact to know whether it should be tested by the jury, defining it as a question of fact if it is tested by the jury adds no information.

The circularity of the functional distinction is not helped by defining the distinction by another function than the one it is being used for. For example, using the allocation between judge and jury in order to determine whether the issue is appealable would not remove the circularity from the definition. It would only make the circularity more indirect. It is very unlikely that the question of whether a certain issue should be allocated to judge or jury would ever be clear, whereas the question of whether the same issue could be appealed is a hard one. If the question of allocation is not clear, answering it brings us back to the circular situation with which we began.

The claim that functional distinction between law and fact is circular is sometimes combined with a claim that it is under- and/or over-inclusive (see for example Mureinik, 1982: 590 f; Thayer, 1890: 156). This objection is problematic. When it is argued that a conception of law or facts is over- or under-inclusive, with what is the conception being compared? Normally, the ground of comparison is the classification made in case law. But for the functional conception, this cannot be the case. The classifications made in case law are already the criteria for classifying something as law or fact. Thus, the over-/under-inclusion objection presupposes the existence of concepts of law and of adjudicated fact other than those employed in the functional conception. It is only possible to claim that it is not only adjudicated facts that are allocated to the jury if one already possesses a concept of fact. ¹⁵

As legal consequences are system-specific, the functional conception will always be system dependent. In other words, legal consequences always belong to a certain legal system, and for that reason a conception of the law/fact distinction making use of those consequences will belong to the same system.

The explanatory potential of extensional definitions

However, Allen and Pardo recognised the weak explanatory potential of the functional distinction, but it is not abandoned because of this. According to their approach there is nothing more than functionality: Talking about ‘law’ and ‘fact’ as two independent concepts just obscures things. On Allen and Pardo’s account the functional distinction add no intentional definition of law and fact. It does not inform us about what it is for something to be of law or of fact, by referring to its function. What it does is to emphasise that there is nothing to law or fact. We cannot say what the questions labelled as facts have in common, as they have nothing in common, except for the fact that they were allocated to the jury. The circularity argument has no bearing on this approach as it dispenses with the conceptual distinction. As the extensional definition does not (here) assume any intentional definition, there is no intentional definition that can be circular.

Kirgis writes, in direct response to Allen and Pardo:

[I]t does not help much to say that functional considerations play an important role in the decision to label a question ‘factual’ or ‘legal’, because the functional considerations are themselves dependent on the characteristics of the matter to be decided. If it is correct that, because of differing competences, there is a set of questions that judges should normally decide and a set of questions that juries should normally decide, then there must be a particular feature about typical adjudicative questions that make them candidates for one or the other set. To understand how functional considerations operate, we need to know the characteristics of the questions that we think judges (or juries) are best suited to answer. Identifying the characteristics of the questions that judge’s label ‘fact’ and ‘law’ is an essential step in understanding the different structural roles of judges and juries. (Kirgis, 2004: 49)

The preference for an intentional definition

The critique of the pure functional conception may be summarised by a quotation from Mureinik: ‘It is not easy to accept that a distinction as conceptual as that between law and fact rests on a foundation so ephemeral and so arbitrary’ (Mureinik, 1982: 590).

The arguments by Allen and Pardo make a good case for concluding that the use of the law/fact distinction in courts today is arbitrary and that it depends on considerations of power distribution rather than conceptual analysis. Their arguments give support to the view that today’s system is not sufficiently protected against this arbitrariness, and it could be argued that there is some evidence for more radical conclusions. As discussed above, Allen and Pardo make a good case for the problems of the majority of the possible distinctions that they discuss. All but the distinction between legal and non-legal fact. As an intentional definition of the law/fact distinction, the distinction between legal and non-legal facts has greater explanatory potential than the functional distinction, Therefore, if I can argue that it is a possible definition, it is preferable.

In the section concerning analytical distinctions in the Myth, Allen and Pardo briefly discuss the distinction between legal and non-legal facts. I will now turn to this promising start, and argue that it can, if elaborated on, make out a solid ground for understanding the law/fact distinction.

The benefit of defining facts negatively

Many articles on the distinction between law and fact, the Myth being one of them, have showed the problems associated with defining a dichotomy between law and fact where both categories are defined as essentially distinct categories in their own capacity. It is difficult, to say the least, to define a dichotomy between law and fact so that both law and fact are defined by sufficient, necessary conditions that are not merely a negative version of the defining condition of the other category. Common for the problematic distinctions are that they do not only try to define what the law is; further, they also strive to define what an issue of fact is on its own virtue, not as not being law. It seems to be this part that is the truly problematic one.

The problem of the law/fact distinction arises when the category of ‘fact’, not in everyday language but as employed in law, is set aside and identified on its own virtue, without taking the law into account. Issues of fact cannot be defined as brute facts, as there are non-legal institutional facts as well (for example the length of a meter). Issues of fact cannot be defined as facts in contrast to norms either, as there are non-legal norms as well (for example norms of etiquette). They cannot be defined as specific as there are general facts as well (for example physical laws), and as Allen and Pardo has showed, they cannot be defined as being objective as the issues of law can be equally objective.

There is a straightforward reason for why the facts are so hard to set aside on their own accord. From the legal perspective, it makes sense to put them together as they are not law. In law, specific legal methods are used to find and identify the content of the law, the judge has extensive education and experience in doing this. As this expertise is delimited to the law, the court cannot rely on legal method or on the proficiency of the judge, for anything except the law. For everything else, evidence must be brought in, expert witnesses may be heard, and reference to common sense might be deployed. Therefore, from the legal perspective, it makes sense to join everything that is not law together in one common category: facts. It is set apart as it is not law. It is hard to imagine that it would be sensible to do this for any other reason than that of the category law already being at the centre.

Many of the distinctions between how matters of law and matters of fact are treated connect to the idea that there is a specific category, the law, of which people trained in that category have better knowledge (Allen and Pardo, 2003a); and where it is possible to learn a specific method for gaining knowledge, the legal method. These ideas lie behind the allocation between judge and jury in the Anglo Saxon legal sphere; the legally schooled judge is presumed to have superior knowledge about law but not about facts. It is also the motivation for the principle of Jura Novit Curia; the judge is considered to have the best knowledge of the law. And in jurisdictions where the right to appeal is restricted to issues of law, this is because judges of higher courts are considered to have better knowledge of law (but not about what is not law). When judges are assumed to have specific knowledge about the law, law is a specific category, a specific topic, of which it is possible to have specific knowledge. Furthermore, when judges are assumed to have better knowledge about law, it is assumed that the question of what the law is is a question of fact, as Allen and Pardo claim.

Let us imagine that a member of the mob is being called to witness at a legal proceeding, in a case against the mafia of which he is a member. According to the law, he is obligated to witness, and to tell the truth, unless he will commit perjury and risk being sentenced. However, according to the code of honour of the mob, he is obligated not to give away any information, to lie about what he knows, unless he will be labelled as a squealer and most likely be punished. Whereas the content of the law is without doubt a question of law, the code of honour of the mafia would be a question of fact if it would be relevant to any legal proceeding. From the perspective of the mobster, there is, however, no reason to believe that the code of honour is something distinct from the law. Rather, the law and the code of honour has a lot in common and it would be reasonable for the mobster to view them as similar and distinguish them from other kind of things, like his own moral view, the probability of being caught and so on.

When introducing the concept of legal and non-legal facts Allen and Pardo state, as quoted above: ‘[The distinction] is one between the facts that the law, by its own conventions, calls “law” and the ones it calls “facts”’ (Allen and Pardo, 2003a: 1805). In this quote lies the greatest benefit of the concepts of legal and non-legal facts: What in law is separated as law and fact are not two different kinds that could be identified without taking the law into consideration. Quite the opposite; the categories of law and fact are legal categories, they are categories created in law, applied in law, onto which the law prescribes different functions.

If this is to be taken seriously, the understanding of facts must somehow relate to law, in its definition of what a ‘fact’ is. By introducing the concept of non-legal facts, the definition of law is conditional for the understanding of fact, that is, of non-legal fact. When the content of law is first known, the content of fact can be learned, as it is what is not law.

What is referred to in court as ‘law’ is the ‘legal fact’, the part of reality identified in law as law. So James B Thayer writes:

All inquiries into the truth, the reality, the actuality of things are inquiries into the fact about them […] But this, it may be said, is a portentous sort of definition; it is turning every question into a question of fact. […] that kind of fact that we call ‘Law’ is discriminated, and set apart under its own name. (Thayer, 1890: 152)

The negative definition of non-legal facts further corresponds to how the distinction is discussed. In the Myth, Allen and Pardo argue that as facts can also be social facts, therefore the law/fact distinction cannot be described in terms of a distinction between natural and social facts (Allen and Pardo, 2003a). Thus, the core element of facts is that they are not law, and therefore any distinction that does not succeed in distinguishing them from law cannot be accepted.

In the process of escaping a problem that has eluded jurisprudence for over a century, we are learning something about the non-legal facts: they are dependent on the category of law. Thus, in order to theorise the distinction between law and fact, we need a theory about the concept of law. There is only so far that we can come without committing ourselves to a theory about the nature of law. Only so far as establishing that the distinction between law and fact is better conceptualised as one between legal and non-legal facts.

The legal aspect of legal facts

An understanding of the law/fact distinction as a distinction between legal and non-legal facts is not complete without a conception of the law. It assumes that when the court is establishing the law it is making use of norm propositions rather than proper norms (Von Wright, 1998: 372). As pointed out by Allen and Pardo, the norm creation by the court is excluded.

According to Allen and Pardo, the norm creation could be seen as factual as well, by analogy with the prediction of factual events. While this might be the case for someone predicting the law when it has not yet been created by the court, and that it is the case appears to be the claim of legal realism, it cannot possibly be the case that the judge is expressing a norm proposition rather than a proper norm when the judge is creating new law. The judge’s creation of a new legal rule can only reasonably be understood as a performative speech act, and to claim that a performative speech act expresses a fact is odd, to say the least.

That the understanding of the law/fact distinction as legal and non-legal facts requires a definition of the law is also a strength of this distinction. The risk of adding essential qualities to the law/fact distinction that is inconsistent with one’s legal theory of choice is less likely when that theory is the basis for the law/fact distinction. Furthermore, as long as the theory of law is universal and account for the differences of legal system, connecting the distinction between law and fact to the definition of law helps to account for differences in the extension of law and fact in different legal systems whilst still building a common ground. Whether something is a legal or a non-legal fact must depend on the content of law of the legal system in question. Only if a something is a part of the relevant legal system is it a legal fact. Whether something is part of the relevant legal system depends on for example, what is pointed out by the rule of recognition of that legal system. The functional distinction is equally flexible between legal systems, but does not have the same explanatory potential when it comes to account for the universal structure of law and fact.

The legal aspect of non-legal facts: more challenges for the distinction

The conceptualisation of the law/fact distinction as the distinction between legal and non-legal fact does not come without problems, however. Allen and Pardo write: ‘Any fact that a juror infers from the evidence presented will depend on whatever general or specific norms the juror possesses or creates (tacitly or explicitly) about what is sufficiently to trigger the inference’ (Allen and Pardo, 2003a: 1805). The quote illustrates vert well what is the problem with the terms ‘legal facts’ and ‘non-legal facts’—or with the term ‘fact’ in the law/fact distinction, for that matter.

If the fact is something for the juror to infer, the fact is something created in the legal process, as a representation of the world; but it is not non-legal in and of itself. As Allen and Pardo point out, the fact that the juror will infer depend on norms, many of which are legal. Beside the norms ‘about what is sufficient to trigger inference’ (the standards of proof), the inference will be affected by norms concerning which evidence is presented for the juror. This point is reaffirmed by Allen and Pardo in their later article, ‘Relative plausibility and its critics’, where they clearly express that fact finding in the legal processes includes ‘some evidentiary rules that pursue values other than truth’ (Allen and Pardo, 2019: 2).

In a legal proceeding, all facts are legal to the extent that the content of any given fact depends, partly, on law. The law is decisive for the selection, description and interpretation of facts. The law, further, determines the sort of evidence on which the fact should be based.

The fact the juror infers may correspond to the state-of-affairs, but it does not have to do so. In this way the term ‘fact’ in the law/fact distinction deviates from how the term ‘fact’ is commonly understood in philosophy. Though there is no consensus on the meaning of the term ‘fact’, ‘fact’ usually refer to the state-of-affairs, ‘what is contingently the case’ or ‘that of which we may have empirical or a posteriori knowledge’. ¹⁶ By contrast, if the fact is inferred from the evidence of the case, the fact cannot be the state-of-affairs, it must be the thought-representation of the state-of-affairs.

Thus, the difference in the meaning of ‘fact’ could be described as a difference in level. The ‘fact’ of the law/fact distinction is a representation of ‘facts’ in the state-of-affair sense. The representation itself is not non-legal; as Allen and Pardo points out, it depends (partly) on (legal) norms, but what it represents is a non-legal state-of-affairs. ¹⁷

If the law/fact distinction is not anchored in a theory of law, then we may only watch, understand and describe them functionally, as Allen and Pardo write. Understanding the law/fact distinction in terms of legal and non-legal facts demands of us to connect the law/fact distinction with the concept of law.

But the terms ‘legal facts’ and ‘non-legal facts’ do not come without problems. The term ‘non-legal fact’ suggests that what we discuss are non-legal states-of-affairs. However, it is not suitable to describe the inference made by the juror as an inference about non-legal state-of-affairs. To describe it as such would force us to conclude that the juror’s inference was incorrect if it did not correspond to the state-of-affairs, but this need not be the case, the juror’s inference might diverge from the state-of-affairs because of the standards of proof. Furthermore, we might well say that the juror’s inference is about the non-legal state-of-affairs, but we may not easily claim that the juror’s inference is non-legal state-of-affairs. We might, however, state that the juror’s inference is the facts of the case. Thus, there is a difference in level between what the juror infers and the non-legal state of affairs. A theory about the ‘facts’ in the law/fact distinction must account for this difference.