Position paper

Metacategories: cross-disciplinary, cross-cultural and cross-temporal perspectives

(Background notes for a workshop to be held at The Australian National University, 18−21 July 2017)

Frances Morphy, Howard Morphy and Jane Simpson


Anthropology and many branches of linguistics have traditionally been concerned with human diversity as well as with attributes that might be potentially universal: they are essentially comparative. There are tensions in both disciplines between approaches that focus on the development of comparative concepts of broad or even universal applicability and approaches that challenge, on theoretical grounds, the applicability of categories and concepts cross-culturally. Certain approaches in contemporary anthropology tend towards the assertion of incommensurability and the existence of many ‘worlds’. Such positions leave open the question of how we translate difference and how people with different ontologies can communicate with one another. Lack of clarity in addressing these tensions inhibits both intra- and interdisciplinary discourse. This workshop is a step towards clarifying the underlying epistemological issues and establishing a discourse that has the potential to bridge theoretical and disciplinary divides across anthropology, linguistics, and other disciplines with similar concerns such as aesthetics, history and philosophy.

Categories and metacategories

Categorisation—the grouping together of different things as instances of the same kind—is an ability possessed by all living creatures, but is especially elaborated in humans, who have language as a tool. A distinction is commonly drawn between the ‘classical’ category and the ‘prototype’ category (Taylor nd: 3ff). The former, which traces its ancestry to the Aristotelian law of the excluded middle, is defined by a conjunction of necessary and sufficient features. Such categories are bounded—an entity is either a member or it is not—and they lack internal structure—all members of a category have equal status as category members. They are also in some sense ideal, permanent, and resistant to change.

The prototype category, in contrast, is defined with respect to a ‘most typical’ core. It has internal structure: some entities are prototypical, others a less than ideal fit. If we take the English category ‘bird’, for example, most English speakers would agree that creatures such as sparrows, robins and hummingbirds are prototypical, whereas penguins, emus and ostriches are not. The prototype category has potentially fuzzy boundaries, and is susceptible to change—or even dissolution—over time. However, most of the research on the internal structures and boundaries of prototype categories has involved experiments on subjects who basically share the same (or very similar) sets of categories (most often psychology students at universities in the Anglosphere). So the question of the universality of particular categories has not been a focus—indeed universality has sometimes just been assumed. If we take the ‘bird’ category for example, it is not self-evident that such a category exists for speakers of Yolŋu-matha, a group of Indigenous languages of north east Arnhem Land. The creatures English-speakers call ‘birds’ are subsumed in the category wäyin, which also includes what English speakers would call land mammals (but not sea mammals).

‘Everyday’ categorisation is mostly implicit; people are socialised into their language’s/culture’s categories and deploy them, for the most part, unconsciously. They appear ‘natural’. This applies as much to practitioners of many academic disciplines as to others. So the study of the process and deployment of categorisation in different societies or languages at different points of time is itself of interest. For example the current argument about marriage in Australia can be framed as a debate between defenders of a classical category—something can only be called ‘marriage’ if it is ‘a union between a man and a woman’—and proponents of a prototype category, who struggle to expand its boundaries to include other types of union.  This is a fierce battle, which shows that categories can matter—they have symbolic value. The defenders of ‘classical’ marriage have long given up saying that other kinds of ‘civil union’ should be illegal, but still they vigorously defend the boundaries of their classical category.

It is also possible in some cases also to trace processes of category formation. For example there is an interesting connection to be drawn between metaphor-making (another seemingly universal human proclivity) and the development of categories of certain kinds. Glucksberg and Keysar (1990) argue that it is a mistake to think of metaphors as invoking a type of comparison, or as implicit similes. Rather, they are ‘class-inclusion assertions’. Taking the metaphor ‘My job is a jail’, they argue that the topic of the metaphor (‘My job’) is assigned to a ‘diagnostic category’. The ‘metaphor vehicle’ (‘a jail’) refers to that category (entities that confine one against one’s will, are unpleasant, and hard to escape from) and is at the same time a prototypical exemplar of it. It is not hard to imagine how ‘frozen metaphors’ might transmute into categories (see Glucksberg 2008).

So possible lines of investigation into human deployment of categories and categorisation could include such questions as: when and under what circumstances are different kinds of categories invoked? How and why do categories change over time? Are there more than two major kinds of categories?

However, we must distinguish between the study of categorisation as embedded in its social/linguistic contexts (see e.g. Bowker & Starr 1999) and the potential use of categorisation as a tool of inquiry into the nature of human beings, societies and languages. Such categorisation would potentially be metacategorisation in at least two of the commonly used senses of meta-: it is meta in the sense that it is, itself, about the categorisation of categories (just as metadata is data about data), and/or it is meta in the sense of being at a higher level of abstraction than categories themselves are (just as metalanguages are specialist tools for describing the properties of languages).

Consciously constructed metacategories are not only a possible tool; they are arguably inescapable since, through using language to describe and generalise about phenomena, we inevitably categorise. We could make a virtue out of necessity. But, possibly, we need to untangle precisely what ‘meta’ we are contemplating when we describe something as a metacategory.

Well formulated metacategories: some preliminary suggestions

Well formulated metacategories could bring apparently disparate phenomena (including apparently disparate categorisations of phenomena) into the same frame so that they can be contrasted and compared.

What might constitute a well formulated metacategory? We advance some preliminary propositions.

  1. Metacategories should be consciously constructed and explicitly articulated, and, above all, useful for comparison cross-culturally and cross-temporally.
  2. They cannot be unexamined prototype categories, because of the risk of essentialising the prototypes of English speakers as universally prototypical. However certain characteristics of prototype categories—fuzzy boundaries, propensity to change through time, and so on, are potentially useful properties.
  3. They cannot be classical categories, because those are essentialist in a different way: they deny the possibility of fuzzy boundaries and of change. It is hard to imagine that any possible useful metacategory takes the form of a classical category. Or is it?
  4. Metacategories are tools for explication rather than for definition or for translation per se. They are arguably a necessary prerequisite for both.

The intersection between the cross-cultural and the cross-temporal becomes immediately apparent in the very process of developing cross-cultural metacategories. We are confronted by the fuzzy nature of many English language categories and the extent to which they change over time. For example contemporary Western English speakers’ conceptualisations of gender and marriage are radically different from those prevailing in the 19th century; hence the comparative perspective from the West at those two points in time would be different. Anthropology has implicitly created atemporal metacategories for a discourse that encompasses different but synergistic ways of thinking about things and acting in the world—art, aesthetics, gender, family, household, law, property, mind, and so on. The cross-cultural metacategory is dialogic. It exists between different local categorisations—be they local in time, or local in space, or both—yet simultaneously strives to encompass difference.

Aims and scope of the workshop

The workshop will address the theme of cross-cultural and cross-temporal metacategories in the context of research into human beings and their societies, languages, and other expressive/communicative forms or actions (art, gesture) from the perspectives of anthropology, aesthetics, linguistics, philosophy and history. The focus will be on issues of explication and translation, and on the epistemological status of metacategories as a tool.

To keep the discussion within manageable bounds, and to focus on topics that have received some attention in all of the aforemetioned disciplines, the workshop will be concerned with the delineation of consciousness and with the expression of properties (such as brilliance, darkness, heaviness) and emotions (such as joy, anger, shame and so on)—thinking of the categorisation of expressive forms or actions, conveyed in gesture or through artworks as well as verbal language.

We will explore the possibility for, and scope of cross-cultural metacategories; some may be universal in scope (such as ‘emotion’ or ‘marriage’, or ‘art’, others regionally or temporally bounded (such as ‘The Dreaming’ in Aboriginal Australia). A main goal of the workshop, which follows from the first, is to improve interdisciplinary communication through this explicitly comparative exercise. In exploring the nature of metacategories we will be also be laying bare the epistemological bases of our respective disciplines, and exploring their similarities and (perhaps irreconcilable) differences.


Bowker, G. C. and Starr, S. L. 1999. Sorting things out: classification and its        consequences. Boston: MIT Press.

Glucksberg, S. 2008. How metaphors create categories—quickly. In R. Gibbs (ed.), Cambridge handbook of metaphor and thought. Cambridge: CUP.

Glucksberg, S. and Keysar, B. 1990. Understanding metaphorical comparisons: beyond similarity. Psychological review, 97(1): 3-18.

Taylor, J. R. nd. Categories and concepts. Undated paper published online at Academia.edu.

  • Australian Government
  • The University of Queensland
  • Australian National University
  • The University of Melbourne
  • Western Sydney University

Subscribe to our newsletter