Farell Ackerman (UCSD)
Explorations into the organization of morphological systems
Linguistics is undergoing an important reconceptualization concerning its methodologies, its objects of inquiry, and ideas about theory construction. In many ways the nature of these changes reflects the fertile influence of older traditions that are often unfamiliar to those trained largely within contemporary varieties of language analysis. Within the newer directions of morphology, there are three interrelated issues that permit the (re)formulation of old questions and guide the formulation of promising hypotheses: 1. (complex) words with functionally discriminable, rather than reliably morphemic, structure are posited as basic objects of morphological analysis, 2. implicative relations are identifiable and quantifiable between words and 3. these relations define emergent cross-linguistic generalizations concerning morphological organization that facilitate learnability.
Linguistics is undergoing an important reconceptualization concerning its methodologies, its objects of inquiry, and ideas about theory construction. In many ways the nature of these changes reflects the fertile influence of older traditions that are often unfamiliar to those trained largely within contemporary varieties of language analysis. Within the newer directions of morphology, there are three interrelated issues that permit the (re)formulation of old questions and guide the formulation of promising hypotheses: 1. (complex) words with functionally discriminable, rather than reliably morphemic, structure are posited as basic objects of morphological analysis, 2. implicative relations are identifiable and quantifiable between words and 3. these relations define emergent cross-linguistic generalizations concerning morphological organization that facilitate learnability.
In a manner to be explored in these lectures, these directions of analysis also align morphological investigation with a parallel resurgence of dynamic systems perspectives on analysis in the developmental sciences exemplified both by (ecological) evolutionary developmental biology and niche construction theory and by variants of developmental systems psychology.
In the broadest terms linguistic and non-linguistic approaches converge in 1. in their emphases on the (probabilistic) modeling of complex co-actions of multi-level systems over time, 2. the crucial roles for system sensitive trajectories of change and 3. the importance of recognizing relations that lead to the emergence of wholes from (configurations of) constitutive parts.
Proposals within Evolutionary Phonology (Juliette Blevins 2004, 2006, 2015; Wedel 2004, 2006, 2007, to appear) can be construed as representing an explicit exploration within phonology of how the role of language development over historical time can be accommodated in accordance with ordinary approaches to understanding natural phenomena. In this connection, encouraging a general application of her approach Blevins (2004: 344) writes:
The evolutionary approach to phonology, morphology, and syntax, is not a theory of what synchronic grammars must encode, but rather what they need not encode as properties of universal grammar. Any cross-linguistic tendency which has a straightforward historical explanation should be excised from statements of universal grammar, unless it can be independently motivated.
Compelling results in phonology from this paradigm have led Anderson 2016: 28 to suggest a quite attenuated role for the Language Faculty and Universal Grammar concerning any plausible appeal to formal and substantive properties:
The tendency in generative treatments of language has been to attribute observed regularities to constraints on the set of possible grammars. Recent work has argued, in contrast, that many if not most such regularities are actually the product of common paths of diachronic change, and thus should be considered regularities in the input data.
In Ackerman and Nikolaeva (2014) this basic approach and its theoretical consequences were investigated at length with respect to the syntax and morphology of prenominal relative clauses. The informing view was that,
Recurrent synchronic grammatical constructions (whether morphological or syntactic, whether widely attested or restricted in distribution) have their origins in diachronic development as potentiated by systemically motivated recombinations of existing elements in several different dimensions of the language system and affected by cognitive, communicative, and use conditions.
In the present series of lectures, under the general label of Word and Pattern (Jim Blevins 2016), I will explore several themes related to a dynamic and developmentally informed approach to complex morphological systems. Among other issues, I will explore the hypothesis that grammar is a complex adaptive system (Ellis et. al. 2008, Ackerman, Blevins, Malouf and Ramscar 2016) in which interactions between its many dimensions and their ingredients produce a canalizing or directing influence concerning what sorts of grammar properties and formal encodings may arise over time. This can be interpreted as a potentiating influence rather than a deterministic one since in many instances structures permitted by particular interactions simply do not occur, though they could have, given different contingent conditions. Thus, certain structures possess an exceedingly low probability of arising, since systemic interactions are unlikely to produce them (Wurzel 1987, Lass 1997, Harris 2008). A Word and Pattern perspective permits a principled examination of the extraordinary variation found in cross-linguistic morphological expression, encouraging inquiry into the nature of “rarity” in language and its relevance for theory construction (Wohlgemuth and Cysouw 2010a, 2010b, Bickel 2015).