Language, Meaning and the Law

In Definition in Theory and Practice (reviewed by the present author in Vol. 63 No. 6 of this journal) Harris and Hutton wrote “semantics is the study and practice of human attempts to impose some degree of communicational determinacy on signs. The successes, failures and limitations of such efforts are, in our view, central to the enterprise of definition“ (p. viii). Their object in that book was to look at the law and lexicography as two longstanding cultural attempts “to impose some degree of communicational determinacy on signs.” In Language, Meaning and the Law, Hutton has continued his investigation into legal language, and this book is, like the earlier, full of observations and arguments that bear not only on the making and interpretation of law but on the theory, construction, use and understanding of ontologies, metadata and search engines. Those who are accustomed to thinking about information as something distinct from language and therefore presenting a separate set of facts and problems will find that assumption increasingly untenable as Hutton's discussion of language in the practice of law unfolds.

Reading the book with my eyes particularly on the lookout for discussions relevant to ontologies, the semantic web and cataloging in libraries – I am, after all, a cataloger – the introduction “Parables of language and law” got my attention immediately.

Babel signifies that the human condition is one of perpetual linguistic and conceptual disorder, and that attempts to establish regimes of cooperation organised through language must struggle to master that disorder. (p. 2).

He immediately follows this statement with a discussion of Wittgenstein's hypothetical “complete primitive language.” He reproduces Wittgenstein's description of that language:

The language is meant to serve for communication between a builder A and an assistant B. A is building with building‐stones: there are blocks, pillars, slabs and beams. B has to pass the stones, and that in the order in which A needs them. For this purpose they use a language consisting of the words “block”, “pillar”, “slab”, “beam”. A calls them out; – B brings the stone which he has learnt to bring at such‐and‐such a call. – Conceive this as a complete primitive language (Wittgenstein, 1978, p. 3, quoted by Hutton on p. 2).

Hutton remarks that “while it might be said to consist of a set of names” this language “is actually a set of commands” (p. 2). There is an unstated but obvious hierarchy and exercise of authority: A speaks, but B, a slave with no voice of her own, obeys. Think of this model in connection with the mechanical slave – the computer, which “knows” only those “blocks”, “pillars”, “slabs” etc. programmed into it – and the computer user who commands the system only as well as s/he knows the language of the program:

The builders' language is an endlessly receycled set of name‐commands, and does not accrue a shared memory of past interactions: it has no history. The builders' world is a totalitarian one in which language, command and obedient act are perfectly coordinated. Therein may lie a clue to its hold over the builders: perhaps authority lies in the language itself and there is no need of law, since in the imaginary domain of this primitive language, there is no room to think outside its categories and therefore no escape from the compulsion it exercises (p. 2).

[…] on being expelled from the God‐given domain of truth and representation, language entered the domain of human responsibility in which “there is no connection between word and world unless we put it there” (Hogan 1996, p. 9). Once it is no longer underwritten by divine authority, the ability of language to depict reality or provide points of shared orientation is constantly in question. There is no single authority which stands outside language and controls it, nor is there a consensus as to how to describe and characterise the world. This raises the question of whether human cooperation through language is only conceivable as a form of coercion, whether it is only through the exercise of power that order can be imposed on the conceptual chaos of language and its relation to the world (p. 5).

Recognizing that we have no gods to guarantee our linguistic cooperation, computer scientists in the 1980s decided that “ontologies” can do the job for us. This soon led Hobbs (1995, p. 819) to claim:

If we are going to have programs that understand language, we will have to encode what words mean. Since words refer to the world, their definitions will have to be in terms of some underlying theory of the world. We will therefore have to construct that theory, and do so in a way that reflects the ontology that is implicit in natural language.

Hobbs offered a brilliantly naive expression of that state of affairs which Hutton described thus: “In the global knowledge economy, the ultimate fetishised object is language itself” (p. 16). Instead of understanding language as an activity in and of the world and being meaningful only as an activity within the world, Hobbs insisted that we must understand the world as “natural language” presents it. We can understand language without understanding or even dealing with the world at all. The reduction of a whole to one of its parts is the precise definition of “fetish” and this fetishization (abstraction) of language (information) from world is the enabling fiction of information technology. Throughout the book Hutton presents the reader with this same enabling fiction as the basis of the operation of law. “The idea of meanings inhereing within texts” he remarks, is one of the fictions about language that allows the interpretation of law through restriction, but “once the processes of law are set in motion, the fictional nature of many of its elements falls from view” (p. 184).

Following the introduction Hutton continues with discussions of various legal issues involving language. Chapter one “Legal theory and Language” looks at discussions of language in theories of natural law, positivism, rule of law, Marxist and critical theories, law and economics and systems theories of law (Luhmann). The second chapter discusses law as descriptive of an autonomous system and as normative prescription, and from there Hutton proceeds through “Philosophy, law and language”, “Issues in legal interpretation”, “Literal meaning, the dictionary and the law”, “Representation, reproduction and intention” and “Idols of the market” before turning, in Part III, Key Issues, to some particular cases. The relevance of chapter two – “Systems theory, normativity and the ‘realist dilemma’” – to information science is particularly obvious, and his discussion of Luhmann pertinent as a bridge between information science, systems theory, social theory and law. In the conclusion, “The semiotics of law, language and money”, Hutton returns to the themes of the introduction, this time with the reader prepared by 175 pages of detailed discussion of a number of divergent opinions on each of the issues addressed.

A description of the central chapters of the book which would link them to the pertinent issues in information science would be nice, but that would require more space that I intend to allow myself here. Instead, I will rejoin him in his conclusion at page 177, where I will follow his own method of misquoting a philosopher to reveal what I (and apparently he) believes to be a crucial similarity between some important social abstractions:

There is no more striking symbol of the completely dynamic character of the world than that of information. The meaning of information lies in the fact that it will be addressed to someone. When information stands still, it is no longer information according to its specific value and significance. The effect that it occasionally exerts in a state of repose arises out of an anticipation of its further motion. Information is nothing but the vehicle for a movement in which everything else that is not in motion is completely extinguished (p. 177).

Now everywhere I have written “information”, Hutton had written “language”. But Hutton was misquoting Simmel (1990, pp. 510‐11), who had written “money” everywhere Hutton wrote “language”. Hutton continues with an exact quotation (actually an English translation) from the same page in Simmel's The Philosophy of Money:

As a tangible item money is the most ephemeral thing in the economic‐practical world; yet in its content it is the most stable, since it stands as a point of indifference and balance between all other phenomena in the world. The ideal purpose of money, as well as of the law, is to be the measure of all things without being measured itself, a purpose that can be realized only by an endless development.

To “the ideal purpose of money, as well as of the law”, Hutton adds language, and I add information. The issue of time – change, movement, memory – looms large when we try to relate an abstraction – money, law, language, information – to social reality. These two issues, time and abstraction,. Hutton brings together in a comment on the history of law:

From the point of view of autonomous legal history, the external, social history of law is “chaos” or “environmental noise”, and is assimilated only with difficulty into the internal history of legal concepts. But even a systems theory of law, as we have seen, must include the dimension of time. The systems view of language as existing outside time is a further step up the ladder towards the Platonic heaven of pure and motionless concepts (p. 181).

Hutton comments on the circulation or motion of signs (money, language, information) and how it comes to the fore in trademark law “since the commercial sign is understood as circulating in the market, a shared product of interaction between the creator‐author, the consumer, media and the public sphere” and that the “law may be required to reduce this movement to a fixed yes‐no determination […] but there is no immediate assumption that all consumers perceive the trademark alike, or that there is a uniformity of reception or meaning” (p. 182). The “single meaning” that law sometimes attempts to determine is a heuristic fiction that the law recognizes as such.

Another issue that Hutton discusses in this context is Google:

Linguistic memory is constantly in motion, utterances are recalled and recontextualised, or simply buried under unbounded layers of accumulated experienced utterances and their contexts. Just as the search engine Google is in a sense bigger than the internet (since it reputedly holds a record of all Google searches), so linguistic memory is bigger than language. More accurately, what we call language is an emergent property of individual and shared linguistic memory […] If a book has millions of pages, an alphabetic index may be of little use. This leads to the conclusion that in relation to the worldwide web and “cyberspace”, what is fundamental not only to memory, but to the creation of a normative order in the present, is the search engine, and other means of tracking, evaluating and organising the massive corpus of information, searches and materials. This is not merely a commercial tool, but also a political and social form, which can reflexively reflect and create different orders of knowledge (p. 180).

Reading this passage here, out of its original context (an extended discussion of the construction, interpretation and contestation of law), it is undoubtedly difficult to see how the passage came to appear in a book on law, what exactly its meaning is, and the implications that may be drawn from it. Yet I hope that readers of this review will recognize that if a book on language and the law leads to such a discussion in its conclusion, then it is relevant to those interested in Google, indexing, the world wide web, search engines “and other means of tracking, evaluating and organising the massive corpus of information” available on the internet.

Following these two discussions, Hutton remarks that “one way of understanding the relationship between legal language and ordinary language is to see that the latter is a fiction created by focusing on the former” (p. 182). To me, this was a relevation, and has an exact bearing on the issue that has been framed in information science as “controlled vocabulary” versus “natural language”. Consider Hutton's claim (actually a conclusion arising from many pages of evidence and argument) in light of the debate within LIS: “natural language” is a fiction created in response to a focus on “controlled vocabulary”, a focus necessitated by the limits and possibilities of several generations of technical systems during the history of information technologies.

Having made this claim – that natural language is an abstraction, i.e. a fiction, made in a particular context – Hutton then proceeds to relate this linguistic matter to a similar abstraction made in law:

When we talk of the autonomy of law, we are recognising its relative imperviousness to direct control by certain social forces, but this requires us to perform an initial conceptual separation between law and society. The autonomy of language is similarly only perceptible if we separate language as system from the totality of human behaviour. Language is none the less frequently conceptualised as a semiotic system which is stable, fixed and still, an ideal reference point. Without such a system, the argument goes, communication would be impossible. But this system can only be activated to serve as a fixed, reference point by putting it in motion, at which point the system of fixed, shared meanings becomes “present to us through its absence” (p. 183).

If social practices are “constituted by fictions […] questions of agency, insight and control over meaning” cannot be avoided: they “haunt the law and legal theory” (p. 183). Here we find the meeting place of law, language, and information science. If lawyers believe that the language of contracts “actually involve “a meeting of minds” (consensus ad idem) as a psychological or social fact, then there are legal theorists who wish to remind them that he world does not come already labeled in “legal categories” and that legal categories do not descend to earth from a Platonic heaven of pure ideas, but are imposed on it by legal analysis” (pp. 183‐4, references omitted). The reader for whom this does not immediately bring to mind the semantic web and related issues of metadata, tagging, cataloging and relevance ranking should probably waste no time on Hutton's book or the rest of this review. For those who do see the connection, I would like to take a short detour through Hutton's earlier work to make a point.

Hutton's first book, Abstraction & Instance, was not a book on legal language, but on semiotic theory, in particular Charles Sanders Pierce's distinction between “type” and “token”. That book was essentially, though not intentionally written as, a profound critique of the foundations of Domanovszky's model of “work”, “expression”, “manifestation” and “item” in his theory of bibliographic description. The issues treated in the first part of the book “involve an exposition and critical analysis of the different illustrations of the type‐token relation made by philosophers and linguists” (p. 3), but neither Domanovszky nor his model are mentioned, and no questions of information science or bibliographical description are ever brought up. In the early chapters Hutton intends to demonstrate “that the intuitive acceptability of the type‐token distinction, and the ease with which it can be illustrated, have led linguists and philosophers to underestimate drastically the complexity of the definitional issues involved” (p. 3). Pierce is himself the best example of the problem Hutton investigates, for in Pierce “the strength of the intuition about the type‐token relation leads Pierce to assert that two tokens of the same type are identical, whereas it is of the nature of the philosophy, and in particular of his theory of signs, to deny the possibility of such an identity” (p. 3). The third chapter illustrates the problem from examples in art theory:

The notion of type is illustrated or suggested by those works of art that can be copied, duplicated, reproduced or forged, such as literary works of art, music, paintings, etc. Again we have an intuitive point of departure, the notion that, for example, two different performances of the same piece of music must in some significant sense be performances of the same work. This intuition serves as the basis for a philosophical discussion of the ultimate nature of works of art. Everyday usage and ways of talking about art are invoked to illustrate the distinction under consideration. Thus we talk of Beethoven's Fifth Sympohony as being a great work, but also of a particular performance being great (p. 4).

Hutton argues that the distinction between type and token is “a useful, but not very profound, distinction between different ways of counting” (p. 5). Types are understood to be “atemporal, non‐material” entities that endow “the physical object, the token, with meaning”, and this involves “an opposition between pure abstraction and pure physical object” (p. 162). The “intuitive” validity of this model lies in the fact that:

It is part of the explicit practice associated with these areas that creative and interpretative work mediates between, for example, score and performance, or script and the stage […]. Further distinctions can be made with respect to the distinction between a book and its copies (p. 163).

Yet Pierce's seemingly “intuitive” notions of type and token (and clearly Domanovszky's work‐expression‐manifestation‐item as well), when erected into a model of the ontology of the existing linguistic, cultural and communicational material, the concrete facts whose meaning and interrelationships we seek to understand, produces insoluble definitional and ontological problems. Abstraction & Instance elucidates those problems but there is barely a hint of political and legal issues, at least not overtly.

Hutton turned to the law at least as early as 1995 with the publication of the paper “Law lessons for linguists? Accountability and acts of professional communication”, and then turned his attention to studying linguistic theory under the Third Reich (Hutton, 1999, 2005). With Harris and Hutton (2007), he returned to language and the law, and in this latest book of his we see the issues of 1990 combined with those of the later 1990s. What unites all these works are “questions of agency, insight and control over meaning” (p. 183) and “accountability and acts of professional communication” (Hutton, 1995). In other words, does language involve anything other than a mechanical matching of tokens with types, or is interpretation involved in all questions of meaning and communication? These questions often seem marginal or irrelevant to the professional linguist and the information scientist, but to one who has spent time studying questions of language and law during the Third Reich, questions of control over meaning and accountability cannot be marginalized for they constitute the central decisions concerning life and death.

The operation of law always involves the interpretation of a preexisting text (law) in order to determine its applicability to a particular case that has arisen after the enactment of the law and therefore not present to the authors of the text of the law. The existence of the law, its interpretation and its application to particular cases all involve some form of coercion or the operation of law would be insignificant. What is of particular interest to Hutton is how that coercion is persistently denied and presented as simply the “facts of language”, whether natural language or the technical language of law, rather than acknowledged as a contingent matter of social relations.

We find an exact parallel in information science. Many recent discussions of the semantic web have offered us a vision of a world in which “meaning” is established unambiguously by url statements in the form of triplets, i.e. syllogisms in the language of IT. The claim that all questions of meaning can be solved (or at least solved “good enough” for machine interpretation) requires that all issues of interpretation, and that means all political, social and legal issues, be banished from consideration and regarded as irrelevant or nonexistent. It is a denial that human communication – via information systems or any other – has any dimension other than a technical operation of matching up signs with meanings (types with their tokens, or works with their expressions, manifestations and items). Such a system of securing meaning through the language of the law has a history much longer than the similar history in information science, and that earlier history is particularly instructive precisely because it was rooted in cultural practices established explicitly for the purpose of and widely believed to be functioning in the interests of the exercise of justice.

The issues in law are identical to the issues in LIS: issues of interpretation (of legal language, laws, cataloging codes and controlled vocabularies), of classifying particular instances (legal cases, documents in a library) in relation to existing precedents (case law, Library of Congress Rule Interpretations), of determining the applicability or non‐applicability of a given law (judicial review, cataloger's judgment), and whether or not an infallible, mechanically applicable rule, hermeneutic, heuristic, tool or algorithm can be developed to eliminate the human variability at the heart of the system (judges, catalogers).

If it is assumed that existing law defines justice (as opposed to the notion that the law must seek to determine what is just in each and every case), a means of interpreting existing law that would be not based on the messy psychology and intellectual limitation of humans but on an infallible interpretative rule would provide us with the perfect system for the administration of justice. Hutton's book reveals, page after page, that no legal hermeneutics of such perfection has ever existed, nor can such infallible interpretive rules or tools ever be developed. Why? Precisely because the exercise of law is the exercise of the power of control over interpretation. Exactly the same applies to information science and the design, construction and use of any and all information systems. To misquote Hutton, “Coercion is the element that ties the names to things, and that coercion is enacted through language as a system of rdf descriptions and urls” (p. 3). To deny the existence of this political matter, masking the operation of power and coercion in matters of interpretation by claiming that some technical system such as the semantic web somehow puts interpretation out of business, is simply a mystification that serves only to prevent us from deepening our understanding of information systems and their operation.

So as we design or dream of future information systems for use in any of our social practices, let us imagine ourselves as the users of the system. Imagine that we are before a judge in a court of law. Is an algorithmic justice good enough? Or would we be demanding an accurate presentation of the facts of the particular case and the circumstances surrounding it, a judge who was neither blindly literal nor accepting whatever interpretation the expert system provided him, and a judicial philosophy that sought to determine how the law should be interpreted and applied in a particular instance to establish justice here and now?

The issues Hutton discusses in Language, Meaning and the Law do not touch upon the trivial and superficial issues related to the design and use of information technologies and may be of little use in dealing with the many of the everyday problems information professionals will encounter. But to understand information systems on the whole as tools used in social practices involving the making and interpretation of linguistic signs, I recommend Language, Meaning and the Law over many other volumes that specifically – and narrowly – address the philosophy and techniques of information science. Anyone who wants to think about the foundations of information science and its relation to social practices will welcome this book. On the other hand, those who imagine that the semantic web will solve all problems of meaning in technical communication will not find it pleasant reading. Unless, of course, they fail to follow the links from law to information science.