Forty Years after Keenan 1976

Subject Properties and Subject Tests

Workshop (7–9 September 2016)


Notifications have been sent. Note that registration for the workshop is free but mandatory. More information regarding registration will be available later.

The distinction between subject and predicate is an ancient one in the history of linguistic thought. It presupposes internal clausal hierarchy, and as such it is, in some sense, a preconception of the modern notion of phrase structure. This early subject concept, however, was not very elaborate, originating from the Classical grammarians. In essence, the nominative was equated with subject, and the subject was equated with nominative (cf. Barðdal 2000). This pretheoretical notion of syntactic relations remained at the core of grammatical description for centuries.

This simple definition of subject was imported into the earliest versions of modern grammatical theory, e.g. in Chomsky’s early discussions of grammatical structure. It was not until the quest for Universal Grammar began that this pretheoretical notion of subject was called into question. Once linguists started describing languages outside of the Indo-European language family, it became evident that subjects behaved differently in different languages. This raised the issue of, for instance, how to distinguish between subjects and topics in languages in general (Li 1976) and what characterizes subjects (Comrie 1973, Anderson 1976, Keenan 1976, Sasse 1978). As a consequence, languages with different alignment of subjects and objects, like accusative, ergative and active languages, began to contribute significantly to the understanding of subjecthood. Keenan (1976), for instance, put forward an exhaustive list of subject properties crosslinguistically, dividing them into the following categories:

  • Coding Properties
  • Behavioral Properties
  • Semantic Properties
  • Pragmatic Properties

Of these, it is first and foremost coding and behavioral properties that have been applied in the linguistic description of subjecthood, with coding properties defined as case marking, agreement and position, while behavioral properties comprise specific syntactic characteristics associated with subjects as opposed to objects. Behavioral properties are, for instance, the ability to control reflexivization and omission on identity in second conjuncts and in controlled infinitives. Since behavioral properties do not require nominative case marking, this led to the recognition that behavioral subjects could be coded in non-canonical cases, for instance in the accusative, dative and genitive (cf. Andrews 1976 on Modern Icelandic and Masica 1976 on the modern South Asian languages). That is, the coding properties discussed by Keenan turn out not to be sufficient criteria for subjecthood, if we adopt a definition of subject based on syntactic criteria instead of only morphosyntactic criteria.

The goal of Keenan’s work was in line with the spirit of the time, namely to identify the properties that define the “universal category” of subject across languages. Contemporaries of Keenan were preoccupied with the particulars of universal categories in language, especially in the wake of Joseph Greenberg’s (1963) universal typology.

In more recent years, there has been a change in focus from universal properties of subjects to language-specific properties of subjects (Dryer 1997, Croft 2001, Culicover & Jackendoff 2005, Van Valin 2005, Bickel 2011). It has become doubtful as to whether or not a universal concept of subject can be maintained because of significant typological differences between languages and the great variety of constructions relevant to this concept; for example, ergative languages code subjects in two separate ways depending on the transitivity of the verb (Queixlós & Gildea 2010, Kikusawa 2002). And even in such closely related languages like Icelandic and German, less than one third of the assumed universal subject properties, as discussed by Keenan, have been regarded as valid tests for both languages (Barðdal 2006: 55, Barðdal, Eythórsson & Dewey 2014). Additional subject tests have even been identified for some languages, like Jónsson (1996) suggests 27 tests for Modern Icelandic.

Other linguistic frameworks have also integrated the concept of subject in different ways into their formalism. In the minimalist framework the subject is defined through the matching of a bundle of interpretable phi- or EPP-features, while in classical government-binding theory (GB) and related theories, subject properties are attributed to a particular functional/structural position in the formal representation of the clause. On optimality-theoretic approaches, the subject is defined through a given set of ranking constraints. In more descriptive frameworks, like Dixon’s (1997: 128–138) Basic Linguistic Theory (cf. also Haspelmath 2004 and Dryer 2006), the subject is simply defined through the subject properties themselves. That is, the argument that passes the subject test is defined as the subject. This descriptive approach is bottom-up, while the theoretical approaches mentioned above are top-down.

A problem with a bottom-up approach to syntactic relations is when a) both arguments of a verb show a subset of the subject properties, e.g. in Dat-Nom constructions where the dative takes on the majority of the behavioral properties but the nominative shows the coding properties, or b) when a subject-like argument shows some of the behavioral properties, but not all. One solution that has been suggested in the literature is to adopt a gradient definition of subject (Ackermann & Moore 2001, Farrell 2010, Serzant 2013). The problem that arises is where exactly to draw the line between the “grades”. In the most radical scenario, the “grades” will be as many as the subject properties, without any principled distinction of which “grades” are more important than others (cf. Croft’s 2001 “methodological opportunism”). It is exactly in such cases where an independent definition of subject may be needed, preferably one that is originally derived from the subject properties in a bottom-up fashion. Problems of this type have not been adequately dealt with in the literature.

The year 2016 marks the 40th anniversary of Edward L. Keenan’s seminal article on subjecthood, “Towards a Universal Definition of Subject”, which came to have an enormous influence on all work on subjecthood for decades after its publication. To celebrate that event, and to review the progress in the field, we launch a call for papers on the the subject concept, the subject properties and the nature and functionality of the subject tests. We invite submissions that contribute to the development of a(n independent) definition of “subject”, including how it is meaningfully distinguished from objects. Topics of interest include, but are not excluded to, the following:

  • behavioral properties of subjects
  • potential behavioral properties of object
  • the waste-paper basket nature of the object category, into which everythingis thrown that does not show  “enough” behavioral properties of subjects
  • language-specificity vs. construction-specificity of the subject, i.e. whetherdifferent subject tests target different types of subject-predicateconstructions within a language
  • whether the tests are real subject tests and not simply morphological case tests
  • linking between syntactic relations and case assignment
  • how different definitions of subject are formalized in different frameworks, including pros and cons of each

Plenary Speakers

  • Spike Gildea (University of Oregon)
  • Jóhannes Gísli Jónsson (University of Iceland)
  • Jóhanna Barðdal (Ghent University)

Deadline for abstract submission: 31 March 2016
Submit (via EasyAbs):


  • Cynthia A. Johnson
  • Esther Le Mair
  • Leonid Kulikov
  • Sigríður Sæunn Sigurðardóttir

EVALISA/Ghent University




  • Ackerman, Farrell & John Moore. 2001. Proto-properties and Grammatical Encoding: A Correspondence Theory of Argument Selection. Stanford: CSLI Publications.
  • Anderson, Stephen R. 1976. On the notion of subject in ergative languages. Subject and Topic, ed. by Charles N. Li, 1–23. New York: Academic Press.
  • Andrews, Avery D. 1976.  The VP complement analysis in Modern Icelandic. Proceedings of the North East Linguistic Society 6, ed. by Alan Ford, John Reighard & Rajendra Singh, 1–21. Amherst: Graduate Linguistics Student Association.
  • Barðdal, Jóhanna. 2000. The subject is nominative! On obsolete axioms and their deep-rootedness. 17th Scandinavian Conference of Linguistics, ed. by Carl-Erik Lindberg & Steffen Nordahl Lund, 93–117. Odense: Institute of Language and Communication.
  • Barðdal, Jóhanna. 2006. Construction-specific properties of syntactic subjects in Icelandic and German. Cognitive Linguistics 17:1.39–106.
  • Barðdal, Jóhanna, Thórhallur Eythórsson & Tonya Kim Dewey. 2014. Alternating Predicates in Icelandic and German: A Sign-Based Construction Grammar Account. Working Papers in Scandinavian Syntax 93: 50–101.
  • Bickel, Balthasar. 2011. Grammatical relations typology. The Oxford Handbook of Language Typology, ed. by Jae Jung Song. 399–444. Oxford: Oxford University Press.
  • Comrie, Bernard. 1973. The Ergative: Variations on a Theme. Lingua 32: 239–253.
  • Croft, William. 2001. Radical Construction Grammar: Syntactic Theory in Typological Perspective. Oxford: Oxford University Press.
  • Culicover, Peter W. & Ray Jackendoff. 2005. Simpler Syntax. Oxford: Oxford University Press.
  • Dixon, R. M. W. 1997. The Rise and Fall of Languages. Cambridge: Cambridge University Press.
  • Dryer, Martin. 2006. Descriptive theories, explanatory theories, and basic linguistic theory. Catching Language: The Standard Challenge of Grammar Writing, ed. by F. Ameka, A. Dench & N. Evans, 207–234. Berlin: Mouton de Gruyter.
  • Dryer, Matthew S. 1997. Are grammatical relations universal? Essays on Language Function and Language Type Dedicated to T. Givón, ed. by Joan Bybee, John Haiman & Sandra A. Thompson, 117–143. Amsterdam: John Benjamins.
  • Farrell, Patrick. 2010. Term Relations and Relational Hierarchies. Hypothesis A/Hypothesis B: Linguistic Explorations in Honor of David M. Perlmutter, ed. by Donna B. Gerdts, John Moore, and Maria Polinsky, 151–172. Cambridge, MA: MIT Press.
  • Greenberg, Joseph H., ed. 1963. Universals of Language. London: MIT Press.
  • Haspelmath, Martin. 2004. Does linguistic explanation presuppose linguistic description? Studies in Language 28.554–79.
  • Jónsson, Jóhannes Gísli. 1996. Clausal Architecture and Case in Icelandic. University of Massachusetts, Amherst, PhD Dissertation.
  • Keenan, Edward L. 1976. Towards a universal definition of subject. Subject and Topic, ed. by Charles N. Li, 303–333. New York: Academic Press.
  • Kikusawa, Ritsuko. 2002. Proto Central Pacific Ergativity: Its Reconstruction and Development in the Fijian, Rotuman and Polynesian languages. Canberra: Pacific Linguistics.     
  • Li, Charles, ed. 1976. Subject and Topic. New York: Academic press.
  • Masica, Colin P. 1976. Defining a Linguistic Area: South Asia. Chicago: University of Chicago Press.
  • Queixalós, Francesc & Spike Gildea. 2010. Manifestations of Ergativity in Amazonia. Ergativity in Amazonia, ed. by Spike Gildea & Francesc Queixalós, 1–25. Amsterdam: John Benjamins.
  • Sasse, Hans-Jürgen. 1978. Subjekt und Ergativ: Zur pragmatischen Grundlage primärer grammatischer Relationen. Folia Linguistica 12.219–252.
  • Seržant, Ilja. 2013. Rise of Canonical Subjecthood. The Diachrony of Non-canonical Subjects, ed. by Ilja A. Seržant & Leonid Kulikov, 283–310. Amsterdam: John Benjamins.
  • Van Valin, Robert D. 2005. Exploring the Syntax–Semantics Interface. Cambridge: Cambridge University Press.