Corpus linguistics and the history of environmental law

The article by Caroline Laske in the new issue of Comparative Legal History, "Corpus linguistics: the digital tool kit for analysing language and the law", argues for the utility of this computational methodology with an example (among others) from the history of environmental law. First the article abstract:

Corpus linguistics methodologies offer innovative ways of reading legal historical sources. Studying the language of source texts using computational techniques that retrieve linguistic data makes detailed searches of words, phrases, and lexical/grammatical patterns and structures possible and provides multiple contextual data that is both quantitative and qualitative, empirical rather than intuitive. It helps us understand not just what is being said, but also how it is being said, how language is used to encode meanings, and what that can tell us about underlying contents and the socio-political, cultural, geopolitical, economic, and other contexts and discourses in which these texts were produced. This paper argues that the use of corpus linguistics is relevant across comparative legal history and can be applied in comparative legal historical research independent of the area of the law or the historical period. Detailed studies incorporating corpus linguistics will be discussed to show the potential of this methodological shift.

The example of environmental law is used to demonstrate how corpus linguistics (CL) can be used to study rapid change in the law (notes omitted):

The general subject of the environment since 1945 can boast an impressive diversity of ideas, concepts, points of view, philosophies, rules, laws, and regulations, to name just a few. As a body of law, its development in a short period of time was governed by rapidity, urgency, and political context. The downside is arguably that the content and legitimacy of the underlying concepts and principles of environmental protection and governance have been given insufficient attention and time to evolve. It is the presence or absence of such legal conceptual underpinnings that will condition whether environmental law develops into a fully-fledged and permanent body of law or whether environmental protection is to be merely a factor in a problem-specific context that will be taken into account in a diversity of established substantive contexts. In the development of what we (rather vaguely) call environmental protection policy and law, we can observe how certain concepts and the semantic meaning of the language used to describe them can differ substantially from one region of the world to another. For example, the concept of ‘person’ is defined differently depending on the legal context. In most civil and common law systems, it relates to a legal entity, an abstraction, or a ‘fiction’, whereas in legal traditions such as indigenous law in Canada or New Zealand, it refers to a real entity of land, such as the legal personality of a river. This illustrates how the understanding of what constitutes the environment and our relationship to it is different depending on the legal context. Revealing such differences can help identify not only law creation and implementation but also inform us of the ontological basis underlying specific legal concepts. In the case of the legal personality of the river, both meanings find expression in positive law. This raises the question of whether and how the positive law can articulate both concepts or whether there is a need to reconsider the concept and its ontological foundation. Moreover, in an international or supranational context, legal concepts and terms can be used differently in different languages, and examining the semantic content can provide a deep-level understanding of such differences, which may be useful, for example, to legal harmonisation projects in EU law.

Studying the language and terminology used in relation to environmental protection and governance diachronically by searching both a general language corpus and an environmental law corpus offers some insight into how these ideas and concepts developed. The underlying premise is again a linguistic pragmatic one, namely, to consider the linguistic expression of and reliable indicators for underlying conceptual thinking. For that purpose, it is interesting to search both a general language corpus as well as a specifically legal one for terms generally associated with environmental protection and governance. A study using the British National Corpus (BNC) and the ECOLEX database showed that the occurrence of language relating to the effects of human activities on the environment has risen exponentially since the late 1970s. This coincides with an increasing preoccupation with environmental protection issues, often sparked and encouraged by major environmental policy events or initiatives to codify environmental principles, such as the UN Conference on the Human Environment in Stockholm in 1972 or the Earth Summit in Rio in 1992, to name two early examples. In this process of rapid expansion of the discourse, the language, and terminology used often evolved by adapting existing terms to new concepts or by creating neologisms.

The BNC is a 100-million-word collection of samples of mainly written language (90%) from a wide range of sources, designed to represent a wide cross-section of British English from the later part of the twentieth century. The corpus is constituted in such a way that it is possible to subdivide it into three separate time periods: 1960–1974, 1975–1984, and 1985–1993. This was particularly useful for retracing the increase of the use of the search terms more precisely. The BNC does not include more recent material and is limited to British English. The results for the search term ‘environment’ are particularly interesting. Its dictionary definition includes several meanings. The context we tend to associate with the term today, namely, that of the effect of human activity on the natural environment, is only one meaning.

This is also reflected in the KWIC [Key-Word-In-Context] lines, of which the first 100 (random order) were examined more closely. During the 1960s and 1970s, the term ‘environment’ was used almost exclusively in a general sense, relating to social, economic, political, and health factors and conditions, while in only a few cases does it referred to the natural/biological/chemical environment. The concept of environment as the effect of human activity on the environment is absent in the texts of that period. The first shift towards an understanding of ‘environment’ as the effect of human activity on the natural environment can be observed between the mid-1970s and mid-1980s, in which three-quarters of the occurrences of the term relate to a general sense and 6% to natural/scientific context. However, we find that in 16% of the lines, environment is used in relation to the effect of human activity on the environment, and 2% refer to titles, such as government departments or ministers, etc. The substantial change can be found in the late 1980s and early 1990s, when the general context uses have fallen to just over a third, the scientific context uses remain a very small fraction, while the uses in relation to the context of the effect of human activity on the environment and specific titles represent over half the occurrences of the term environment. This linguistic evidence of general language usage suggests it was during this particular time that the concept of environment took on and consolidated the specific meaning relating to the effect of human activity on the environment, an understanding that is most common today. It is also interesting to note that the general uses show a shift in preoccupation from one period to the other. In the 1960–1974 subcorpus KWIC lines, environment would refer almost exclusively to the conditions of social, work, and children’s education. This gives way in the 1975–1984 subcorpus to uses in relation to health and safety in relation to domestic, hospital/patient, and road user situations. The context that is added in the 1985–1993 subcorpus are environments in relation to digital technologies, such as specific system or network environments.

The equivalent study based on legal texts was more complex, as there is no appropriate environmental law texts corpus readily available. Like the study on consideration described above, a purpose-built corpus had to be constituted. The choice of database fell on ECOLEX, which is an information service on environmental law operated jointly by FAO, IUCN, and UNEP. Its purpose is to build capacity worldwide by providing the most comprehensive possible global source of information on environmental law. The ECOLEX database includes information on treaties, international soft-law and other non-binding policy and technical guidance documents, national legislation, judicial decisions, and law and policy literature. Users have direct access to the abstracts and indexing information about each document, as well as to the full text of most of the information provided. However, access to the full texts is given through links – the database does not host as such a corpus of texts. The choice was made to concentrate on the texts listed under case law/jurisprudence in the ECOLEX database, which include case reports from national jurisdictions worldwide, as well as from international courts such as the European Court of Human Rights or the Court of Justice of the European Union. To the extent that we are dealing with reports of court cases, including, at times, opinions (eg, from judges, attorney general), these represent naturally occurring language and provide good insight into language usages. The corpus constituted for the study contains 1,486,541 words.

In contrast to the multitude of meanings that could be found for terms in the BNC, such as environments or ecology, the textual context that the ECOLEX corpus provided showed the language and terminology semantically restricted to the issue raised by the impact of the human activity on the environment. All terms examined in this study collocated and clustered with typical environmental protection terminology: eg, protect/protection, health, effects or impact, water, land, noise, nitrates, and agriculture. Also prominent are terms relating to the political, administrative, and legal structures, etc, set up as part of the environment protection policy: eg, department, minister, planning, management, and law.

Examining the textual context revealed the evolution of the language and meanings in relation to newly emerging concepts of environmental law. Some terms have traditionally been associated with environmental issues, such as contamination, environment, fishing, land, negligence, nuisance, pollution, preservation, waste, and water. More recently, other terms appeared, such as biodiversity, ecology, ecosystem, and organic; and some may express concepts that have been discussed in relation to the environment only during the last few decades: animal welfare, climate (change), ecocide, human rights, (social) justice, sustainability, and sustainable development. Unsurprisingly, the more common terms of environmental law are also among those that show the highest frequency, such as environment, land, pollution, waste, and water. Three notable exceptions that have commonly been associated with the early environment case law in the common law are nuisance, negligence, and contamination. It is some indication that the approach of litigating early environmental contamination cases through the common law of nuisance and negligence did not persist on a worldwide level.... The figures would no doubt be a little different if carried out on a corpus that only contained texts of common law cases. Similarly, other legal traditions such as indigenous law in the US, Canada, or New Zealand may place different emphasis and/or introduce other concepts, such as ‘Mother Earth’. For example, the cluster for ‘water’ would be the concept of ‘Zaagidowin’ (love/taking-care-of), the principle of harmony and healing. The general increase in both the occurrences and diversity of the terms shows an important expansion of environmental law language and terminology. Long-established preoccupation with issues relating, for example, to land, pollution, or water, continue to be discussed and are not superseded by newer concepts.

I have to say that I'm not convinced that all this advances our historical understanding of environmental law very much beyond what we already know. Do you disagree? Please comment.