Given the absence of grammaticality judgments concerning the relevant data, isn't historical syntax just a necessarily hobbled version of synchronic syntax, and can’t everything that we want to learn from historical syntax be more easily and reliably learned from synchronic data?
Ellegård, Alvar. 1953. The auxiliary do: The establishment and regulation of its use in English. Gothenburg Studies in English. Stockholm: Almqvist and Wiksell. Kroch, Anthony. 1989. Reflexes of grammar in patterns of language change. Language variation and change 1:3, 199-244.
Kroch, Anthony. 1994. Morphosyntactic variation. In K. Beals et al., eds., Proceedings of the Thirtieth Annual Meeting of the Chicago Linguistic Society, vol. 2. Chicago: University of Chicago Press. 180–201.
Kroch, Anthony. 2001. Syntactic change. In Mark Baltin and Chris Collins, eds., The handbook of contemporary syntactic theory. Malden, MA: Blackwell. 629-739.
But there’s a bright side: In contrast to grammaticality judgments, the corpus data that forms the basis of research in historical syntax includes additional information - namely, information concerning the frequency of variants in usage. Moreover, historical corpus data contain information concerning the time course of the various frequencies.The proper quantitative analysis of historical corpus data involves figuring out how to combine the diachronic frequency information with conventional grammatical analysis.
Kroch, Anthony. 2013. Modeling language change and language acquisition. Ms., University of Pennsylvania.
A reasonable way of modeling the S-shaped frequency trajectories is by
assuming that they reflect the logistic, a mathematical function with
two parameters: a slope (steepness) and an intercept. There are now
many studies showing that when we are dealing with a single grammatical
change, the slope parameter is constant across different contexts (the
Constant Rate Effect). In other words, even though particular contexts
might favor or disfavor a change, the effect of the contexts is
orthogonal to the rate of change itself (statistical independence).
Kroch, Anthony.
1995.
The statistical modeling of language change.
Ms., University of Pennsylvania.
Santorini, Beatrice.
1993.
The
rate of phrase structure change in the history of Yiddish.
Language variation and change 5, 257-283.
Taylor, Ann.
1994.
The change from VO to OV in Ancient Greek.
Language variation and change 6, 1-37.
Annotation
guidelines for PPCHE. These guidelines form the basis for the
guidelines adopted by related projects.
Annotation guidelines for historical French.
Installing
Linux on Windows with Ubuntu.
CorpusStudio (another way of running CorpusSearch
on a Windows machine).
GLEEFUL
2014: Searching parsed corpora with CorpusSearch
LSA Summer
Institute 2013 Workshop on using parsed corpora for research in
historical syntax
3. What is the relationship of the frequency information and the
grammatical analysis? (top)
How are the variants in the texts to be characterized grammatically?
One possibility is that an old grammar allows two variants, with one of
the variants driving out the other. At the end of the transition, the
winning form is reanalyzed as the result of a new grammar.
Alternatively, the appearance of the eventually winning form might
itself be taken as reflecting the emergence of the new grammar. In this
case, there isn’t necessarily a reanalysis, and the transition period
reflects competition between two competing grammars, which ends with the
loss of the one of the grammars. There may be other scenarios. For
instance, what about cases where a particular grammatical option only
wins out in some contexts, but not others (potentially relevant in
connection with relative clause extraposition)?
4. How can we mathematically model diachronic change? (top)
Relevant evidence concerning these two (and perhaps other) scenarios
comes from a close analysis of the time course of changes. In
particular, the trajectory of changes (at least, completed ones) follows
an S-shaped curve. We also find such curves in contexts of biological
mutation, where organisms with a fitness-enhancing mutation replace
their less fit counterparts.
5. Extracting data from parsed corpora (top)
Overview of CorpusSearch. Discussion of ordinary queries, coding
queries, and revision queries. For examples of queries, check out the
Dropbox folder for the course (specifically the CorpusSearch_queries
directory with the following subdirectories):
Penn Parsed Corpora of
Historical English (PPCHE) (with links to other parsed historical corpora).
6. Headedness in VP (the change from OV to VO) (top)
In principle, an OV grammar can generate VO orders and vice versa. Is
it possible to tell which grammars underlie texts during periods of
transition from OV to VO? We investigate the question based on data on
Old French and Old Italian.
Kroch, Anthony and Beatrice Santorini.
2017.
Detecting grammatical properties in usage data.
39th annual meeting of the
Deutsche Gesellschaft für Sprachwissenschaft,
Universität des Saarlandes.
7. Participle agreement (top)
According to the Académie Française, participle agreement
is optional in examples like les lettres que j'ai
écrit(es) 'the letters that I wrote (lit. have written)'. We
investigate the decline of gender and number agreement in the history of
French and perhaps also Old Italian and explore whether the decline is
related to the phrase structure change from OV to VO.
Kroch, Anthony and Beatrice Santorini.
2014.
On the word order of Early Old French.
SinFonIJA 7, Graz.
8. Relative clause extraposition (top)
Time permitting, we explore the slow decline of relative clause
extraposition in various languages for which we have historical
corpora and the issues that arise concerning statistical independence.
Last revised: 19 Jul 2017
Wallenberg, Joel.
2016.
Extraposition is
disappearing.
Language 92, e237-e256.