Radical Pro Drop and Fusional Pronominal Morphology in Colloquial Singapore English: Reply to Neeleman and Szendrői

1 Neeleman and Szendrői's (2007) Radical-Pro-Drop Generalization

Neeleman and Szendrői (2007) (N&S) propose that radical pro drop (RPD), namely, the liberal omission of any grammatical argument in languages like Chinese, requires agglutinating morphology on pronouns. N&S derive this generalization from three assumptions: (a) null arguments are zero spell-outs of regular pronouns (Perlmutter 1971); (b) spell-out rules for pronouns may target nonterminal nodes in the syntax (Weerman and Evers-Vermeul 2002); and (c) the Elsewhere Principle (Kiparsky 1973) obtains, with its three notable features shown in (1a-c).

1. 1.

   1. a. All else being equal, a phonological realization of a category C takes priority over a phonological realization of the categories contained in C.

   2. b. All else being equal, a phonological realization of a category C that spells out more of C's features takes priority over a phonological realization that spells out fewer features.

   3. c. Optionality results if the phonological realization of a category C spells out fewer of C's features than the phonological realization of the categories contained in C.

(N&S 2007:687)

Using English and Japanese, I illustrate here how the RPD generalization is derived. Suppose that RPD is due to the spell-out rule in (2). English pronouns are fusional for case, while Japanese pronouns are agglutinating. This difference is captured by rules for him and kare in (3) and (4a-b), respectively.¹ N&S use the features [+p(ronominal), -a(naphoric)] to indicate that K(ase)P is a pronoun (p. 682).

1. 2. [_KP +p, -a] ↔ Ø

(N&S 2007:682)

1. 3. [_KP +p, -a, 3, SG, M, ACC] ↔ /him/

(cf. N&S 2007:687)

1. 4.

   1. a. [_NP +p, -a, 3, SG, M] ↔ /kare/

   2. b. [_K ACC] ↔ /o/

(N&S 2007:688)

RPD is unavailable in English because the Elsewhere Principle always prefers overt spell-out rules such as (3) over the zero spell-out rule in (2). Consider (5) for English pronouns.

1. 5.

(cf. N&S 2007:688)

(2) and (3) both target the KP. (3) blocks (2) due to (1b) because (3) realizes more features (i.e., Case and φ-features) than (2). Now, compare (5) with (6) for Japanese pronouns.

1. 6.

(cf. N&S 2007:688)

(2) and (4a) do not compete in (6). (2) is more compliant with (1a) than (4a), but (4a) is more compliant with (1b) than (2). As a result, neither rule blocks the other. Thus, Japanese emerges as an RPD language. Two notes are in order. First, N&S's analysis allows for the possibility that languages with fusional pronominal morphology have some version of the pro-drop option. Spanish and Italian allow (subject) drop. This is captured by the context-sensitive rule in (7).

1. 7. [_KP +p, -a, φ_i] ↔ Ø /_____ [φ_i]

(N&S 2007:687)

(7) is not in an elsewhere relation with rules like (3): (7) contains agreement that (3) lacks, whereas (3) mentions particular φ-features that (7) is insensitive to. Second, N&S's analysis predicts that RPD is possible in a language as long as its pronominal paradigm has some KP-internal agglutinating morphology (case, number, or some other nominal features). Thus, Chinese allows RPD, N&S argue, because plural pronouns are derived from singular variants by the plural morpheme men, as illustrated in (8a-b).

1. 8.

   1. a. [_NP +p, -a, 1, SG] ↔ /wŏ/

   2. b. [PL] ↔ /men/

(N&S 2007:689)

N&S show that the RPD generalization is crosslinguistically robust by testing it against a sample of twenty languages and The World Atlas of Language Structures (Haspelmath et al. 2005). In this squib, I present data from Colloquial Singapore English (CSE) that question the robustness of the RPD generalization. In section 2, I provide examples showing that CSE is an RPD language and point out that this variety has fusional pronominal morphology, just like Standard English. These results indicate that the RPD option is not necessarily conditioned by the agglutinative morphology on pronouns. I also review Saito's (2007) generalization that RPD is made possible by lack of agreement...