Academic journal article Memory & Cognition

Familiarity and Relational Preference in the Understanding of Noun-Noun Compounds

Academic journal article Memory & Cognition

Familiarity and Relational Preference in the Understanding of Noun-Noun Compounds

Article excerpt

When people are presented with noun-noun compounds, they tend to produce two main types of interpretation: relational and property interpretations. One theory of compounding maintains that relational interpretations are preferred over property ones. However, many of the studies supporting this relational preference hypothesis appear to be vitiated by a failure to control for the familiarity of compounds. A rating study on so-called "novel" compounds used in previous studies is reported, which shows that many can be considered to be familiar. Then two experiments in which familiarity is controlled are presented, to test the relational preference hypothesis, using a sensibility judgment task (Experiment 1) and a comprehension judgment task (Experiment 2). The results show that familiarity has a clear effect on the ease of understanding of noun-noun compounds but that there is no hard evidence for relational preference. The implications of these results for the empirical literature and for current theories are discussed.

Combinations of concepts are regularly used in everyday life to coin new names for new objects and to extend the boundaries of our language (Jespersen, 1942). Indeed, most of the new phrases entering the language each year are made from combinations of existing words (e.g., laptop computer, video player, and soccer mom). As new compounds enter a language, they are understood by the language community, exhibiting some of the fundamental generativity of natural language (see, e.g., Costello & Keane, 1997,2000,2001 ; Gagné, 2000; Lynott, Tagalakis, & Keane, 2004; Tagalakis & Keane, 2004; Wisniewski, 1996). Typically, it has been found that even apparently simple novel compounds admit great combinatorial richness in their semantics; for example, the compound finger cup could be "a cup to clean fingers in," "a thin cup that is the width of a finger," "a cup shaped like a finger," and so on. The attempt to explain these and other phenomena are the main drivers of empirical and theoretical research into conceptual combination.

In this article, two key aspects of conceptual combination that provide us with important empirical leverage in distinguishing between different models of combination are examined: relational preference and familiarity. As will be shown, the resolution of these two issues has significant implications for our theoretical understanding of conceptual combination and the adequacy of past empirical work.

Relational Preference

In recent years, there has been considerable controversy about whether people have a preference for some interpretations of novel compounds over others (see Costello & Keane, 1997; Gagné, 2000; Wisniewski, 1996; Wisniewski & Centner, 1991 ; Wisniewski & Love, 1998). Two main types of interpretation have been identified in the diversity of meanings that people produce for novel compounds: relational and property interpretations. Relational interpretations hinge on a relation being found to connect the two concepts (e.g., "a cup to clean fingers in," where to-clean-in is the relation that connects finger and cup; see Murphy, 1988, 2002). Property interpretations hinge on asserting a property that belongs to one concept of the other concept (e.g., "a thin cup," where the property thin is used from finger, see Wisniewski & Centner, 1991). Recently, it has been argued that relational interpretations are fas preferred interpretations and that property interpretations are considered only when no relational ones can be found (Gagné & Shoben, 1997; Shoben & Gagné, 1997). As Shoben and Gagné claimed, "We do not dispute that property matches occur; however, we contend that property matches are the interpretation of last resort" (p. 35). We call this view the relational preference hypothesis.

Evidence for Relational Preference

The relational preference hypothesis is supported by a variety of studies in which different measures have been used. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.