What could have sanctioned null subjects in Old English?

Chapter 5 investigates the question of what could have sanctioned null subjects in Old English. It tests the empirical validity for the Old English data of central syntactic models attempting to account for the appearance of null subjects. It is shown that the predictions made by these analyses are not borne out in Old English, and it is argued that analysing the omitted subjects in terms of ellipsis is more fruitful than construing them as an active canonical or partial pro-drop grammar. The stance is taken that referential null subjects do not represent a productive grammatical feature of Old English, and it is again noted that the occurrence could instead be considered linguistic ‘residue’. Such an analysis would not be at odds with work which analyses pro-drop as argument ellipsis cross-linguistically.

The morphosyntactic characteristics of null subjects in Old English

Chapter 4 provides an in-depth quantitative investigation of the morpohsyntactic characteristics of null subjects in Old English. The distribution of overt and null subjects is presented according to the structural variables of clause type, the position of the finite verb, person, and number. Attempts are made to fit an explanatory generalized mixed-effects logistic regression model incorporating these linguistic variables, as well as those non-linguistic variables which emerged as significant in Chapter 3. It is demonstrated that correlations between the occurrence of null subjects and such variables are extremely weak when both genre and the individual text are taken into account: only minuscule influence on the odds of having a null subject instead of an overt one is exercised by these variables. It is argued that this strengthens the view of null subjects in Old English as linguistic ‘residue’.

The long-term diachrony of referential null subjects in early English

Chapter 6 investigates the long-term diachrony of null subjects in early English. On the basis of a large-scale empirical analysis of overt and null referential subjects in Old, Middle, and Early Modern English, it is argued that no real diachronic decline in the possibility of null subjects can be detected across a period spanning c.850 years of early English texts. It is argued that this is consonant with a story where Old English does not feature a productive pro-drop property. The chapter also shows that verb-initial clausal syntax and conjunct clause environments constitute the strongest favouring effects for null subjects in Middle and Early Modern English, as is also the case in Old English. An argument is made that this is consonant with a story where subject omission in early English is viewed as a form of argument ellipsis.

Introduction

This chapter introduces the issues to be investigated in the book. It defines the concept of referential null subject, and gives an outline of previous statements concerning the status of such subjects in Old English, as well as a presentation of the null subject phenomenon as it manifests itself in the wider Old Germanic context. Previous works on null subjects in Old English have reached widely diverging conclusions: some scholars say that null subjects did not occur in Old English, while others state that Old English was a canonical pro-drop language. Walkden (2013) suggested the compromise that null subjects are a case of diatopic variation. The question of the possibility of null subjects in Old English, and the diverging descriptions and explanations of this phenomenon, are presented as the main motivations for the book. The chapter also introduces the data material and methods used.

Conclusions

Chapter 7 concludes the book by giving a summary of the background for the study, as well as of its main results. The chapter makes note of the widely diverging previous descriptions of null subjects in Old English, and concludes that this language was not a canonical pro-drop language, instead suggesting that null subjects are linguistic 'residue'. Moreover, it argues that there is no evidence for a ‘dialect split’ with regard to subject omission. As certain conditions that are often taken to license null subjects do not appear to be present in Old English, it is suggested that the omitted subjects may be analysed as argument ellipsis. These findings are corroborated by the Middle and Early Modern English data, which display similarities to Old English with regard to the scarcity of null subjects, as well as to their morphosyntactic characteristics.

Do Anglian dialects of Old English have a partial pro-drop property?

Chapter 3 deals with the question of whether Anglian dialects of Old English, in contrast to the West Saxon literary standard, had a partial pro-drop property. The chapter investigates this ‘dialect-split hypothesis’ by means of descriptive statistics and inferential statistical modelling. It is also noted that what has been interpreted as diatopic variation could also be representative of other types of variation, and consequently the variables of translation status, period, and genre are also investigated, in addition to dialect. The primary analytical techniques used in this chapter are generalized fixed-effects logistic regression modelling and random forests of conditional inference trees. The chapter concludes that the dialect-split hypothesis must be considered falsified.

Referential null subjects in Old English

Chapter 2 investigates the question of whether Old English was a canonical pro-drop language. It does so by giving tabulated contrastive overviews of the occurrence of null subjects in 181 Old English prose and verse texts. It is demonstrated that null subjects occur extremely rarely in Old English prose, but that they are much more frequent in the verse. On the basis of this, the chapter argues that Old English was not a canonical pro-drop language. Instead, it is suggested that null subjects may represent linguistic ‘residue’. It is acknowledged that the view that Old English is a canonical pro-drop language is understandable in light of the higher proportions of null subjects in the poetry, but the case is made that evidence from the poetry cannot be taken to be syntactically representative of Old English, either quantitatively or qualitatively.