Deconstructing the Internal/External Causation Distinction

Malka Rappaport Hovav, Hebrew University of Jerusalem

The distinction between internal and external causation was introduced by Levin and Rappaport Hovav (1995) and further developed by Harley and Noyer (2000) and Alexiadou, Anagnastopoulou and Schäfer (2006, 2015) to account for the distribution of verbs in the variants of the causative alternation and to account for patterns of morphological markings in causative alternation.   LRH stress that events in the world are not classified as internally or externally caused, but rather verbs representing construals of events.   Verbs participating in the causative alternation are basically dyadic and the alternation comes about by the lexical binding of the external argument.  Verbs of internal causation are lexically monadic and do not participate in the causative alternation because the binding of the external argument is not relevant for these verbs.  Alexiadou and Alexiadou et. al. classify roots as: internally caused, externally caused, agentive or cause-unspecified, with causative alternation verbs being understood as cause-unspecified.  However, the distinctions have been intuitively applied and no clear semantic explication which can be characterized independently of the phenomenon the distinctions are supposed to account for has been offered.

In a recent study of the causative alternation in English (Rappaport Hovav 2014), I suggest that there is no grammatically relevant distinction between roots or verbs in terms of type of causation.  In that study, I argued that all causative alternation verbs are basically monadic.   I argued that there are general nonlexical constraints governing the (non)appearance of an external argument which take into consideration the discourse context of the verb.  There is a general constraint that a lexical causative must express direct causation and furthermore the (non) appearance of the external argument is governed by Gricean maxims.

This talk further extends these ideas, taking into account the problems raised by Neeleman and van de Koot (2012) for the common explication of the notion of direct causation.  I suggest that most causative alternation verbs are based on roots denoting states.  Depending on the argument the state is predicated of, it will be construed as a state with a propensity to change or a propensity not to change.  This is the crucial (nonlexical) distinction which enters into the calculation of whether the external argument can, cannot or must be expressed in the description of particular events of change of state.  The particular choice of cause for lexical causatives is argued to be determined not by the length of the causal chain but rather by whether the links in the causal chain between the chosen cause and the change of state can be seen as following the natural course of events.


Alexiadou, A., Anagnostopoulou, E., Schäfer, F., 2006. “The properties of anticausatives cross-linguistically,” In: Frascarelli, M. Ed., Phases of Interpretation. Mouton, Berlin.

Alexiadou, A., Anagnostopoulou, E., Schäfer, F., 2015. External Arguments in Transitivity Altnernations: A Layering Approach. Oxford University Press.

Harley and Noyer, R. 2000. “Licensing in the non-lexicalist lexicon,” In Peeters B., Ed., The Lexicon/Encyclopaedia Interface,  Amsterdam: Elsevier Press.

Levin, B. and Rappaport Hovav, M. 1995. Unaccusativity, Cambridge, MA: MIT Press.

Neeleman,A. and van de Koot, H. 2012. “The Linguistic expression of causation,” in M. Everaert, M. Marelj and T. Siloni, Eds. The Theta-System, Oxford University Press.

Rappaport Hovav, M. 2014. “Lexical content and context: The causative alternation in English revisited,” Lingua 141.