Null and cognate objects and changes in (in)transitivity

Abstract

This paper examines the historical development of null objects (e.g., *He took the loaf and held 0 / between his hands) and cognate objects (e.g., He sang a song) in English. We will demonstrate that English lost definite/referential (and indefinite) null objects (only generic null objects are possible in present-day English, e.g., They have the ability to impress and delight 0 /) but extended the range of cognate objects (which is now also possible with activity/event nouns, e.g., He smiled a disarming smile). Cross-linguistic diachronic comparison (using data from the history of the Greek language) reveals that the historical development described is language-specific and that both directions of change (loss or rise) of null and cognate objects are possible. We will test the hypothesis that there is a connection between the availability of such (de)transitivization processes and changes that affect the syntactic representation of aspectual distinctions. More precisely, we will examine the hypothesis that in English, both the loss of definite null objects and the rise of cognate objects with activity/event nouns are linked to the grammaticalization of the viewpoint (progressive) aspect (be + V + ing). We will show that the grammaticalization of the viewpoint aspect in English has not affected the development and loss of null object constructions. However, the grammaticalization of the viewpoint aspect appears to have progressed in parallel with activity/event-noun cognate objects.

The diachrony of the cognate object construction in Romanian and Hungarian

“…But crucially, aspectual COCs are only found in diverse translations of the Holy Bible, see (21a), or in poetry, see (21b), as the everyday language blocks them almost completely, see 21c Instead, in the absence of aspectual COCs, where the object is literally cognate to the prototypical unergative verb it accompanies, Hungarian has accusative-marked (-t) pseudo-objects (POs) (Piñón, 2001;É. Kiss, 2004;Kiefer, 2006;Csirmaz, 2008;Halm, 2012), that is, non-subcategorized, non-thematic and non-referential nominals, which take on the role of the aspectual CO in the language, as convincingly demonstrated and argued in Farkas (2019Farkas ( , 2020a. More precisely, these latter studies show that POs should be classified into three, where the motivation behind the following ternary division-instead of a binary one, which merges the POs of class (b) and (c)-is syntactically supported 2 .…”

“…Building on the cross-linguistic observation generally valid in the majority of languages according to which cognate objects (COs) and cognate object constructions (COCs) are frequently subject to diachronic changes (see Horrocks & Stavrou, 2010or Lavidas, 2013b Greek; Mittwoch, 1998 for Hebrew;Visser, 1963, Lavidas, 2013a, 2018or van Gelderen, 2018 for English), this article offers an account of language change in these objects and these constructions in the history of Romanian and Hungarian. While it examines and compares the changes affecting them, the paper shows that these constructions in the two typologically unrelated languages have undergone diachronic change in two different directions: whereas the former language is characterized by a loss of the COC, the latter one is characterized by an increased use of the (especially aspectual) COC.…”

“…(2) a. Mary slept a fitful slumber. b. they rejoiced with exceeding great joy (kjv,Matt,2,10) According to their most recent classification (see Horrocks & Stavrou, 2010;Lavidas, 2013aLavidas, ,b, 2014Lavidas, , 2018, these constructions are of three types. First, a transitive COC is built on a prototypical transitive verb, which selects a subject and obligatorily an object as well as in to do deeds.…”

“…These cognate nominals are only accidentally morphological COs of the verb, they do not denote events, they are fully referential arguments, hence they do not share the properties of aspectual COs (see below); cf. Mittwoch (1998), Pereltsvaig (2002), de Swart (2007, Horrocks & Stavrou (2010), and Lavidas (2013a). Third, an aspectual COC is built on a prototypical unergative verb, which requires only a subject and, exceptionally, can also take an accusative nominal, which is exclusively a cognate object as in to sleep a sound sleep.…”

“…In addition, the main aim of this comparative analysis is not, first and foremost, to contribute to understanding Romanian as a foreign language by a native speaker of Hungarian (or vice versa) or to gaining insights into the complexity of Romanian through Hungarian (or vice versa). But the primary and intended purpose of the present analysis is mainly to contribute to the cross-linguistic observation concerning the diachronic changes of COCs, whereby Romanian is similar to (Ancient) Greek (Horrocks & Stavrou, 2010;Lavidas, 2013b) and Hungarian is comparable to English (Visser, 1963;Lavidas, 2013aLavidas, , 2018; van Gelderen, 2018). However, this similarity is not purely coincidental and is also mirrored, for instance, in the morphosyntactic properties of events of transition in these languages.…”