Grammatical variation and complexity

In the last two decades, research interest in grammatical variation and the determinants affecting variation has been constantly growing. Research in this area has emphasized the role of complexity as an important, if not the most important variable driving linguistic variation in many areas. undefinedRohdenburg (1996), for example, proposed the Complexity Principle which argues that there is a trade-off relation between the cognitive complexity of the contextual environment and the grammatical explicitness:

Complexity principle
In the case of more or less explicit grammatical options the more explicit one(s) will tend to be favoured in cognitively more complex enfironments

In other words, if speakers can choose between two grammatical variants, they will prefer the more explicit variant in cognitively more complex environments.

Among others, this proposal is taken up by undefinedMondorf (2009) to account for the choice between analytic comparatives (e.g. more wealthy) and synthetic comparatives (e.g. wealthier). Due to the one-to-one mapping of a grammatical function to each word, the analytic comparative is considered to be more explicit than the synthetic comparative in which two functions are mapped onto a single morphologically complex word. undefinedMondorf (2009: 6) proposes the principle of analytic support:

Analytic support
In cognitively more demanding environments which require an increased processing load, language users – when faced with the option between a synthetic and analytic variant – tend to compensate for the additional effort by resorting to the analytic form.

Thus, adjectives which are more difficult to process should occur more frequently with the analytic comparatives, and a similar claim can be made for other areas of the English grammar in which both an analytic and a synthetic variant are available, as with the possessive variation between analytic of possessives and synthetic 's possessives. Following the Complexity Principle, nouns with higher processing complexity should show a stronger preference for the of possessive than nouns with lower processing complexity.

An empricial challenge

However, the evidence for this effect of processing complexity on the choice of grammatical variants is almost exclusively indirect. undefinedMondorf (2009), for example, argues that the factors that increase the probability of analytic comparatives (e.g. the length of the adjective, the number of morphemes, or whether it is in attributive or predicative position) all increase the cognitive complexity either of the adjective itself or of the environment in which the adjective occurs. Yet, there is no independent evidence that these factors do indeed have an increasing effect on the processing complexity of the adjective. Similarly, undefinedHinrichs & Szmrecsanyi (2007) relate some the factors that increase the probability of of possessives to syntactic complexity and to processing difficulties, but there is no direct evidence that analytic possessives are indeed preferred if the processing load increases.

Despite the substantial progress that has been made in researching the link between grammatical variation and processing complexity, the mostly indirect evidence in support of the Complexity Principle presented in the literature can be seen as indicative of how little is known about the psycholinguistic reality of the effect of processing complexity on the preference of analytic forms (yet, see undefinedBoyd 2007 on comparative alternation as a notable exception). In line with this view, undefinedRosenbach (2014: 242) emphasizes that “research into genitive variation needs to be connected more closely to psycho-/neurolinguistic evidence and models”. From a more general perspective, undefinedSzmrecsanyi (2009: 322) remarks that the link between analyticity, syntheticity and language complexity “is not always backed up by hard empirical evidence, especially when it comes to processing complexity”.

Obtaining some answers

This research project provides some “hard empirical evidence” (undefinedSzmrecsanyi 2009: 322) by examining the proposed relation between the preference of analytic constructions and processing complexity in comparative and possessive alternation in English.


A series of production and perception experiments were conducted to seek psycholinguistic support for the assumption that analytic constructions do have a processing advantage over corresponding synthetic equivalents. Corpus studies using large data sets from the BNC and COCA complement the experimental data. Thus, the project tested the hypothesis that the analytic variant is available to speakers as a means of compensating for an increased processing complexity. By looking at two independent types of grammatical variation, comparatives and possessives, the results address the more fundamental question of whether a compensatory mechanism can be assumed to be at work in English morphosyntax in general.

Results and implications

A central result is that analytic forms do not have an advantage for the listener in perceptual processing. On the contrary, synthetic comparatives as well as combinations of existing bases with illegal suffixes are processed significantly faster than the corresponding analytic comparatives. This strongly suggests that morphological processing in perception is much more efficient than syntactic processing. However, the experiments and corpus studies reported in the book provide strong evidence that language users are sensitive to complexity effects in a way that is compatible with the Complexity Principle in production: the analytic comparative of adjectives that are difficult to process is significantly more probable than the synthetic comparative. The fact that the effect of processing complexity in production is more strongly visible for comparatives than for possessives suggests that there are other determinants of morphosyntactic variation that can override the contribution of complexity.

Taken together, the results imply that analyticity and syntheticity both have their own advantages: the first in production, the latter in perception. This distinction between production complexity and perceptual complexity is often overlooked in the literature, but it may contribute to the recent discussion on conflicts in linguistic motivation as presented, for example, in undefinedMacWhinney, Malchukov & Moravcsik (2014). In particular, it adds a dimension to Mondorf’s contribution on competing motivations in English morphosyntactic variation in that volume.


Boyd, Jeremy. 2007. Comparatively speaking. A psycholinguistic study of optionality in grammar. Ph.D. dissertation, University of California, San Diego.

Hinrichs, Lars & Benedikt Szmrecsanyi. 2007. Recent changes in the function and frequency of Standard English genitive constructions: a multivariate analysis of tagged corpora. English Language and Linguistics 11(3), 437–474.

Mondorf, Britta. 2009. More support for more-support. Amsterdam: John Benjamins.

Rohdenburg, Günter. 1996. Cognitive complexity and increased grammatical explicitness in English. Cognitive Linguistics 7(2), 149–182.

Rosenbach, Anette. 2014. English genitive variation – the state of the art. English Language and Linguistics 18(02), 215–262.

Szmrecsanyi, Benedikt. 2009. Typological parameters of intralingual variability: Grammatical analyticity versus syntheticity in varieties of English. Language Variation and Change 21(3), 319–353.

MacWhinney, Brian, Andrej Malchukov and Edith Moravcsik (eds.). 2014. Competing motivations in grammar and usage. Oxford: Oxford University Press.

Responsible for the content: E-MailWebredaktion English Language and Linguistics