Elsevier

Cognition

Volume 109, Issue 2, November 2008, Pages 235-250
Cognition

The cross-linguistic categorization of everyday events: A study of cutting and breaking

https://doi.org/10.1016/j.cognition.2008.08.009Get rights and content

Abstract

The cross-linguistic investigation of semantic categories has a long history, spanning many disciplines and covering many domains. But the extent to which semantic categories are universal or language-specific remains highly controversial. Focusing on the domain of events involving material destruction (“cutting and breaking” events, for short), this study investigates how speakers of different languages implicitly categorize such events through the verbs they use to talk about them. Speakers of 28 typologically, genetically and geographically diverse languages were asked to describe the events shown in a set of videoclips, and the distribution of their verbs across the events was analyzed with multivariate statistics. The results show that there is considerable agreement across languages in the dimensions along which cutting and breaking events are distinguished, although there is variation in the number of categories and the placement of their boundaries. This suggests that there are strong constraints in human event categorization, and that variation is played out within a restricted semantic space.

Introduction

The systematic cross-linguistic study of how semantic categories are expressed in signs (words, morphemes, constructions, etc.) has a long history spanning many disciplines, including anthropology, linguistics and psychology (for recent overviews see Boster, 2005, Evans, in press, Koptjevskaja-Tamm et al., 2007). Classic domains of enquiry include color (Berlin & Kay, 1969), emotion (Ekman, 1972, Ekman and Friesen, 1975), ethnobiology (Berlin, 1992, Berlin et al., 1973), the human body (Andersen, 1978, Brown, 1976) and kinship (Goodenough, 1956, Lounsbury, 1956). Despite this work, there is still little consensus on the degree to which categorization is constrained across languages by general principles or is relatively free to vary. Universals of nomenclature once thought to have been established are now being reexamined in the light of new empirical data – cf. Roberson, Davies, and Davidoff (2000) versus Kay and Regier (2003) on color; Russell (1994) versus Ekman (1994) on emotion; Majid, Enfield, and van Staden (2006) versus Wierzbicka (2007) on the body. The outcomes are still to be determined.

A major advance in this debate has been the insistence on collecting data in a standardized way across a wide range of languages and cultures, along with the use of statistical techniques to quantify the extent of agreement across languages (Croft and Poole, 2008, Kay and Regier, 2003, Levinson and Meira, 2003, Regier et al., 2007). In this paper, we follow this approach in an investigation of the semantic categorization of events of “cutting and breaking”. We examine how people describe a standardized set of such events across a typologically, genetically and geographically diverse set of languages, focusing in particular on the implicit categorization of events imposed by the verbs speakers use. We show that despite variation in the number of “cutting and breaking” categories recognized in the different languages, and in the exact boundaries of these categories, there is considerable common structure in the data set: the languages share a semantic space which can be characterized by a small set of dimensions. We argue that these dimensions reveal a common conceptualization of “cutting and breaking” events.

Categorization is the process by which different entities – objects, events, relationships, properties, etc. – are treated as being “of the same kind” for the purposes of language, memory and reasoning. Thus, a poodle, a snake and an octopus, although they are perceptually quite dissimilar, can be grouped together under the category of “animals”.

Although categorization has been studied in a number of semantic domains, the primary focus among psychologists has for many decades been on the representation of objects (Murphy, 2002). Comparatively little work has been done on event categorization (although there has been a noticeable awakening of interest in events in recent years; see the edited volumes by Hirsh-Pasek and Golinkoff, 2006, Shipley and Zacks, 2008, Tomasello and Merriman, 1995). Just as for objects, however, we can ask whether different happenings are viewed as events of the same kind. For instance, is slicing a carrot with a knife the same kind of event as cutting a piece of paper with a pair of scissors (e.g., both “cutting” events)? What about using an axe to chop a branch in two versus smashing a pot with a hammer? And – a critical question – are the answers to such questions largely universal, or do they vary from language to language?

Questions like these have rarely been pursued. Many studies of the linguistic structuring of events have focused on how global meaning elements are characteristically “packaged”– i.e., distributed across the lexical items of a sentence. For example, motion events can be broken down into a number of structural components like the fact of motion, the manner of motion, the path along which the motion takes place and the nature of the moving object. Languages vary in which of these elements are typically expressed in the verb (Talmy, 1985). In English, for example, the verb characteristically expresses both the fact of motion and the manner of motion, with path information expressed in a separate element such as a particle (e.g., dance in, walk out, stroll over). In Spanish, in contrast, the verb typically combines information about motion with information about path, and expresses manner optionally in an adverbial (e.g., entrar bailando ‘go-in dancing’; salir caminando ‘go-out walking’). Differences like these have been studied extensively from the standpoint of both language acquisition (e.g., Allen et al., 2007, Choi and Bowerman, 1991, Choi et al., 1999, Zheng and Goldin-Meadow, 2002) and the cognitive consequences for event construal and memory (Finkbeiner et al., 2002, Gennari et al., 2002, Kita and Özyürek, 2003, Papafragou et al., 2002, Papafragou et al., 2006).

In studies of motion events, cross-linguistic similarities and differences in categorization are seldom investigated (see Choi & Bowerman, 1991, for an exception). Rather, it is commonly assumed that categories of path, manner, etc. are largely universal, with cross-linguistic variation revolving around how the elements expressing them are separated or combined in lexical items. There is, however, one recent study that directly addresses categorization for a subtype of motion events, human locomotion (Malt et al., 2008).

Humans have two efficient locomotion gaits – walking and running (Alexander, 1992, Minetti and Alexander, 1997). These gaits can be characterized by a number of parameters such as stride length, the length of time that each foot is on the ground and the time course of the forces exerted on the ground. Walking involves a pendulum-like motion from the hip, where one foot is always in contact with the ground, whereas running involves a bounce-and-recoil motion in which there is a moment when neither foot is in contact with the ground. As speed of bipedal locomotion increases, there is an abrupt transition from walking to running (Diedrich & Warren, 1995). Malt et al. (2008) investigated whether this discontinuity in human locomotion constrains linguistic categorization. That is, they asked whether languages universally honor this discontinuity, categorizing human locomotion according to the gait involved, or instead categorize locomotion in different ways according to language-specific principles.

Japanese, Spanish, Dutch and English speakers were shown videoclips of a woman on a treadmill and asked to describe what she was doing. In all the languages a major distinction was made between the two sorts of gaits, i.e., naming responses showed a categorical distribution for walking and running. This was true for both manner-salient languages like English and Dutch as well as path-(and ground-)salient languages like Spanish and Japanese (see Slobin, 2004, on this distinction). This study demonstrates that languages can share principles of fine-grained event categorization even though they may differ in how they package the components of a motion event.

In this paper, we also investigate fine-grained linguistic event categorization, using a new technique and applying it to a new domain: everyday events involving a “separation in the material integrity” of an object (Hale & Keyser’s, 1987, term), or “cutting and breaking” events, for short.1 In their investigation of locomotion Malt et al. could rely on previous biomechanical analyses in order to hypothesize possible cross-linguistic constraints on gait-naming. But no such analysis exists for events like slicing a carrot with a knife or chopping a branch with an axe. Instead, our research attempts to uncover which parameters are relevant to the categorization of such events.

The domain of “cutting and breaking” was chosen in part because its cognitive status is ambiguous: a priori, it seemed equally plausible that event categories in this domain are universal and that they are variable. In favor of universality, the manufacture and use of tools for purposes of cutting and breaking has been dated back at least 2.5 million years to the East African Rift area. Modern humans (Homo sapiens sapiens) appear to be unique in making and using tools especially for “cutting”, such as pressure-flaked knives (de Beaune, 2004, Harris, 1983, Toth and Schick, 1993).2 Cutting and breaking are practiced in every society and by practically every member of the society; these actions do not require specialized knowledge (although there may, of course, be expert variants of cutting and breaking in a community, such as diamond cutting and quarrying). The fact that cutting and breaking have been central to human activity for so long suggests that there may be a common way of conceiving such events.

Further reason to expect universality comes from linguistics, where cross-linguistic analyses suggest that verbs of cutting and breaking fall universally into two distinct classes, which have systematically different kinds of meanings that correlate with distinct syntactic behaviors (Guerssel et al., 1985, Levin and Rappaport, 1995; but see Bohnemeyer, 2007). These claims suggest that there is a shared human conceptualization of cutting and breaking events, at least at a relatively abstract level.

But there is also evidence for variability in the conceptualization of these events. First is the fact that although cutting and breaking behaviors have been part of the human repertoire since prehistory, their exact manifestation varies according to the particular ecology and practices of a community. For instance, Americans and Europeans chop vegetables by holding them still and bringing a knife down on them from above, whereas Punjabi speakers in rural Pakistan and India often move the vegetables against a stationary curved knife. Different practices like these could lead to systematic differences in people’s event categories.

Consistent with this view, the extensions of “cutting and breaking” verbs have been claimed to differ considerably across languages (Fujii, 1999, Goddard and Wierzbicka, in press, Pye, 1996, Pye et al., 1995). For example, English speakers use break for actions on a wide range of objects (e.g., a plate, a stick, a rope), while speakers of K’iche’ Maya must choose from among a large set of “breaking” verbs on the basis of properties of the object; e.g., -paxi:j ‘break a rock, glass, or clay thing’ (e.g., a plate); -q’upi:j ‘break (another kind of) hard thing’ (e.g., a stick); -t’oqopi:j ‘break a long flexible thing’ (e.g., a rope) (Pye, 1996, Pye et al., 1995). Cross-linguistic variation suggests that no one way to categorize “cutting and breaking” events is cognitively obvious or inevitable.

As further evidence for flexibility in the human categorization of “cutting and breaking” events, young children make many errors in verb choice in this domain (Bowerman, 2005, Pye et al., 1995, Schaefer, 1979). In English such errors include, for example, saying break for tearing cloth and cut for pulling apart a peach slice with the fingers or crushing ice cubes with a rolling pin. Learners of English also often overextend break/broken to reversible events like opening a safety pin, undoing overall straps, or separating magnets, and open to irreversible events like pulling off a doll’s leg or breaking a roll (Bowerman, 2005). This suggests that the boundary between reversible and irreversible separation events might not be honored as systematically in all languages as it is in English.

The existing cross-linguistic evidence on semantic categorization in the domain of cutting and breaking is piecemeal, and limited to a few languages. In the present study, we tackle the question of universality versus relativity in this domain systematically by examining the lexical categories employed by speakers of a wide range of diverse languages in describing a standardized set of events. If certain distinctions or groupings recur across a wide range of languages, it is plausible to assume that these reflect conceptualizations that are fundamental to human cognition.

Section snippets

Participants

Event descriptions were collected from speakers of 28 typologically, genetically and geographically diverse languages, drawn from 23 countries, 13 language families and a range of cultures (see Table 1). For each language there were between one and seven consultants (M = 3.25). The 24 researchers listed in Table 1 collaborated in this effort; all were experts on the language they studied.

Materials

The data were collected using a set of 61 videoclips depicting a wide range of events (Bohnemeyer, Bowerman, &

Results

Speakers’ event descriptions can be treated as analogous to data obtained in sorting tasks designed to study categorization (Bowerman, 1996). In a typical sorting task, a participant might receive a set of cards, each depicting a different item, and be asked to sort the cards into groups of similar items. Speakers in the present study were not asked to sort, but rather simply to describe what they saw in the videoclips. But each different verb they applied to the target events was taken to

Discussion

Speakers of typologically, genetically and geographically diverse languages show considerable agreement in how they implicitly categorize events of “cutting and breaking” through the verbs they use to describe them. First, they agree on treating such events as a relatively coherent semantic domain. At the beginning of this investigation it was unclear whether there is such a thing as a core semantic domain of “cutting and breaking” events –i.e., whether languages routinely distinguish these

Acknowledgements

This study of the semantic categories of “cutting and breaking” events took place in the Event Representation project at the Max Planck Institute for Psycholinguistics. We thank all our colleagues who contributed their insights, data and analyses to the study. Special thanks to Steve Levinson and Leah Roberts for comments on earlier drafts. The research was supported by the Max Planck Gesellschaft, as well as by a European Union Marie Curie Fellowship awarded to the first author. The authors

References (81)

  • M. Zheng et al.

    Thought before language: How deaf and hearing children express motion events across cultures

    Cognition

    (2002)
  • R.McN. Alexander

    A model of bipedal locomotion on compliant legs

    Philosophical Transactions of the Royal Society of London

    (1992)
  • E.S. Andersen

    Lexical universals of body-part terminology

  • S.A. de Beaune

    The invention of technology: Prehistory and cognition

    Current Anthropology

    (2004)
  • B. Berlin

    Ethnobiological classification: Principles of categorization of plants and animals in traditional societies

    (1992)
  • B. Berlin et al.

    General principles of classification and nomenclature in folk biology

    American Anthropologist

    (1973)
  • B. Berlin et al.

    Basic color terms: Their universality and evolution

    (1969)
  • D. Bickerton

    Roots of language

    (1981)
  • P. Bloom

    How children learn the meanings of words

    (2000)
  • J. Bohnemeyer

    Morpholexical transparency and the argument structure of verbs of cutting and breaking

    Cognitive Linguistics

    (2007)
  • J. Bohnemeyer et al.

    Cut and break clips, version 3

  • J.S. Boster

    Categories and cognitive anthropology

  • J.S. Boster et al.

    Form or function: A comparison of expert and novice judgments of similarity among fish

    American Anthropologist

    (1989)
  • M. Bowerman

    Learning how to structure space for language: A crosslinguistic perspective

  • M. Bowerman

    Why can’t you ‘open’ a nut or ‘break’ a cooked noodle? Learning covert object categories in action word meanings

  • M. Bowerman et al.

    Shaping meanings for language: Universal and language-specific in the aquisition of spatial semantic categories

  • C.H. Brown

    General principles of human anatomical partonomy and speculations on the growth of partonomic nomenclature

    American Ethnologist

    (1976)
  • P. Brown

    ‘She had just cut/broken off her head’: Cutting and breaking verbs in Tzeltal

    Cognitive Linguistics

    (2007)
  • M. Casasola et al.

    Can English-learning toddlers acquire and generalize a novel spatial word?

    First Language

    (2006)
  • W. Croft et al.

    Inferring universals from grammatical variation: Multidimensional scaling for typological analysis

    Theoretical Linguistics

    (2008)
  • F.J. Diedrich et al.

    Why change gaits? Dynamics of the walk-run transition

    Journal of Experimental Psychology: Human Perception and Performance

    (1995)
  • P. Ekman

    Univerals and cultural differences in facial expressions of emotion

  • P. Ekman

    Strong evidence for universals in facial expressions: A reply to Russell’s mistaken critique

    Psychological Bulletin

    (1994)
  • P. Ekman et al.

    Unmasking the face

    (1975)
  • Evans, N. (in press). Semantic typology. In J. J. Song (Ed.), The handbook of linguistic typology. Oxford: Oxford...
  • M. Finkbeiner et al.

    The role of language in memory for actions

    Journal of Psycholinguistic Research

    (2002)
  • Y. Fujii

    The story of “break”: Cognitive categories of objects and the system of verbs

  • D. Gentner

    Some interesting differences between nouns and verbs

    Cognition and Brain Theory

    (1981)
  • D. Gentner

    Why nouns are learned before verbs: Linguistic relativity versus natural partitioning

  • D. Gentner et al.

    Individuation, relativity, and early word learning

  • View full text