Sank's Glossary of Linguistics 
ComplexC-ComplexR

* COMPLEXCODA

  1. (Optimality Theory) A markedness constraint:
    * COMPLEXCoda
    Codas cannot be complex.
     | Janina Mołczanow, 2020
  2. (Optimality Theory) A markedness constraint:
    * COMPLEXCODA
    One violation per every segment in a coda beyond one (i.e. codas are allowed to consist of at most one segment).
     | Joanna Chociej, 2009
  3. (Optimality Theory) A markedness constraint:
    * COMPLEXCODA (Prince and Smolensky (1993 / 2004)
    Syllables must not have complex codas.
     | Dylan Herrick, 1999
  4. (Example)
     ○ Ajluni Arabic (Afro-Asiatic; Jordan) does not permit coda clusters because the * COMPLEXcoda constraint over-ranks the SON constraint.
    /qalb/ * COMPLEXcoda SON DEP-V-IO * Coda
    ☞/qalib/  *  *
     /qalb/  *!  *
     | Hana Asaad Daana, Maisa Sadi Jaber, and Sereen Jubran, 2023
     ○ 
    1. Serbo-Croatian (Western South Slavic; Serbia, Croatia, Bosnia and Herzegovina, Montenegro)
    Input:
    /kovert/
    MAX-STEM *COMPLEX
    CODA
    DEP *[HIGH] *[LOW]
     a. [kovert]  *!
    ☞b. [koverat]  *  *
     c. [kover]  *!
     d. [kove]  *!*
     e. [koverit]  *  *!
     Tableau 1 shows the evaluation through which candidate (b) emerges as the winner. The most faithful candidate, which is (a), is ruled out because it fatally violates *COMPLEXCODA, which is quite high in the hierarchy. Candidate (c) avoids the violation of *COMPLEXCODA by deleting the last consonant, but incurs a violation of MAX-STEM, the highest-ranking constraint. Candidate (d) dispenses with the coda cluster altogether, thus violating MAX-STEM twice. Both candidates (b) and (e) violate DEP, but while (b) inserts a low vowel in order to disrupt the cluster, (e) inserts a high one. Since *[HIGH] is ranked higher than *[LOW], candidate (b) is the winner. | Marko Simonović and Antonio Baroni, 2014
     ○ In the initial state of the child's grammar, all markedness constraints are ranked above the faithfulness constraint (Gnanadesikan 1995). Faithfulness then gradually rises in the hierarchy, overtaking the markedness constraints one by one: first *CODA, then ONSET, then (variably) *COMPLEXCODA or *COMPLEXONSET, and finally the remaining one. In the end, faithfulness is ranked on top and the child masters all syllable structures. | Paul Boersma and Clara Levelt, 1999

COMPLEXITY-BASED ORDERING

  1. (Morphology) The Complexity-Based Ordering Hypothesis (CBO) (Hay 2003/2004) posits that cognitive processing restricts affix combinability above and beyond structural restrictions (e.g. inter-nation-al-iz-ation, but *internation-iz-ation-al, despite being interpretable and meeting structural restrictions). Specifically, CBO states that affixes that tend to be parsed during lexical access occur farther from the root (in English) than affixes that do not, because this facilitates lexical access. | Andrea Sims and Jeffrey Parker, 2012
  2. (Morphology) Jennifer Hay (2002) proposed a psycholinguistic approach to affix ordering now known as Complexity-Based Ordering (CBO), which claims that affix order is determined by the parsability of the affixes, i.e. more separable affixes can appear only outside of less separable affixes. Hay shows that this principle accounts for why many grammatical affix combinations are unattested. CBO has since been supported by research of derivational affixes (English prefixes, English suffixes and Russian suffixes). | Robert Reynolds, 2013
  3. (Morphology) A psycholinguistic model of morphological complexity, according to which an affix which can be easily parsed out in processing should not occur inside an affix which cannot. This model has been called complexity based ordering. The general claim is that affixes can be approximately ordered along a hierarchy of complexity, with more separable affixes at one end, and less separable affixes at the other end. More separable affixes can attach outside less separable affixes, but not vice-versa. | Jennifer Hay and Ingo Flag, 2004
  4. (Morphology) In addition to selectional restrictions, Hay (2002, 2003/2004) proposed a general processing constraint, named complexity-based ordering by Plag (2002), that limits the number of possible combinations of affixes. The underlying idea is that the morphological separability of affix and base is a graded phenomenon, with far-reaching consequences for affix stacking:
    While some affixes basically tolerate no internal structure, others will tolerate structure to some minimum degree. The degree of internal structure tolerated by an affix is ... determined ... by how much structure that affix, itself, creates. Phrased in terms of processing, an affix that can be easily parsed out should not occur inside an affix that cannot. (Hay 2002)
     | Ingo Plag and Harald Baayen, 2009

COMPLEXITY FILTER

  1. (Syntax) In Koopman and Szabolcsi (2000) we account for the different word order and constituency of verbal complexes in Hungarian, Dutch and German by a derivational theory which relies on overt (remnant) XP movement only. The derivations yield paradigms that are not always attested in full in a particular language: there are language-specific gaps in the paradigms. Koopman and Szabolcsi argue that these gaps should be accounted for by a new brand of filters, complexity filters, which act on the representations that the derivations generate. Complexity filters are sensitive to overt material only, and impose restrictions on the "size" or "internal complexity" of certain constituents in designated Spec positions at the end of the derivation.
     Movement is not subject to economy conditions: the computational system is blind and fully automatic. Different surface patterns result from the particular history of the derivation: which parts of the structure may be split (expressed by language specific "splitting" parameters), and what "size" constituent is allowed to occupy a particular Spec position at the end of the derivation (expressed by language-specific complexity filters). Splitting parameters and complexity filters are independent, but in many cases interact to restrict what size constituent can be pied-piped, and what parts of the structure muse be "chopped up" into smaller pieces. | Hilda Koopman, 2008
  2. (Semantics) Katzir (2007) and Fox and Katzir (2011) formalize a notion of complexity that requires formal alternatives to be no more complex than the asserted one. This formal complexity filter on Excl can be stated thusly:
    Complexity Filter on Excl
    Excl(φ) ⊆ {Basic(ψ): ψ is no more complex than φ}
     | Bernhard Schwarz and Michael Wagner, 2024
  3. (Examples)
     ○ To see how arbitrary the Koopman and Szabolcsi (2000) provisions for Hungarian are, consider how the language would look if we simply changed one of them, and kept the others the same. For example, suppose the Hungarian complexity filter forced the movement of XPs smaller than InflP, rather than larger than InflP. The language that results from such an adjustment would have a bizarre collection of properties form the point of view of the accounts which make the X⁰ distinction. But isn't that a point in favor of such theories? | Edwin Williams, 2004
     ○ Fanselow and Ćavar (2001) argue that the German paradigm (1) shows the need for complexity restrictions independent of head status. Verbs pied-pipe their unstressed particles when they undergo V2 movement (1a,1c), while stressed particles are stranded (1b,1d). (1) indeed establishes the need for a morpho-phonological complexity filter for the second position. (1) also shows that lexical entries can be split up in a V2 construction. (1) does not show that elements other than an X⁰ category can occupy the second position, however. The paradigm in (1) constitutes no reason for abandoning the idea that X⁰ elements only undergo V2 movement in German.

    1. a.
      b.
       
      c.
      d.
      dass
      dass
      that
      er
      er
      er
      er
      he
      beginnt
      fängt
      den
      den
      the
      den
      den
      Brief
      Brief
      letter
      Brief
      Brief
      beginnt
      an.fängt
      begins

      an




        '(that) he begins with the letter'

     | Gisbert Fanselow, 2004

COMPLEXITY THEORY

  1. (Acquisition) Chaos / complexity theory was first brought to the attention of second-language acquisition researchers in the US with the pathbreaking article by Diane Larsen-Freeman (1997) that offered an alternative to the simplistic input-output, linear, information processing model of second language acquisition. It has been followed by a host of other studies that flesh out general principles of complexity / dynamic systems theory as applied to SLA (see Larsen-Freeman 2020, de Bot et al. 2007, Larsen-Freeman and Cameron 2008, Larsen-Freeman 2011, a.o.). In 1997, Larsen-Freeman spelled out principles of a chaos / complexity theory for language learning, elaborating on some of them in 2011:

    1. Complex systems (like the weather, economic systems, human learning) are open and dynamic.
      Complex "systems" have no distinct boundaries; they exist only because of the fluxes that feed them, and they disappear in the absence of such fluxes. One could therefore say that a complex system is dynamic rather than static; it exists only in the interaction between things and is therefore not itself a thing (Osberg 2008).
    2. Complex systems operate under conditions that are not in equilibrium.
      When you learn one additional piece of knowledge, this new knowledge doesn't just add itself to the other things you acquired previously. The equilibrium you thought you had reached in your prior state of knowledge gets disrupted as one new piece of knowledge reconfigures the whole picture.
    3. Change / dynamism are central: The systems adapt both through interaction with the environment and through internal reorganization / self-organization.
      Complex systems are systems because they are comprised of many elements or agents, which interact. Despite the linear nature of textbooks and the sequential order of items on the syllabus, learning does not grow in chapters or units that one can check off, test, and consider "acquired". Each new item presented in a new chapter is related to an item learned earlier, but on a more complex plane, because one can now illuminate the item from more angles in light of other items acquired since then. This most important principle of complexity theory has been glossed generally in the following way: "The act of playing the game has a way of changing the rules" (Gleick 1987/2008).
    4. The strength of the interactions changes over time: Therefore, there are often multiple routes possible between components, mediated in different ways.
      Unlike learning historical or mathematical facts, learning how to communicate in a second language is variously successful within an individual's biography and among individuals. complexity theory looks at the whole ecology of learning: the learner in interaction with current others (teacher, textbook, fellow learners, native speakers), with absent or with past others (through texts), with his / her perceptions of present and past others, of past and present selves, and with whole discourses about the language, its speakers, its writers and the ideologies and worldviews they vehiculate.
    5. The complexity of complex systems is emergent: It is not built into any one element or agent, but rather arises from their interaction.
      Through soft assembly and co-adaptation, patterns emerge or self-organize. As the biologist Sandra Mitchell writes: "Self-organization refers to any set of processes in which order emerges from the interaction of the components of system without direction from external factors and without a plan of the order embedded in an individual component" (Mitchell 2003). These patterns or performance stabilities are transformed with further usage. In other words, you don't learn the rules and then apply them; instead, the rule emerges out of the interaction itself.

     | Claire Kramsch, 2012
  2. (Acquisition) The introduction of Complex Dynamic Systems Theory (CDST) into the realm of applied linguistics has attracted attention to the unpredictability and uniqueness of second language development among learners (Larsen-Freeman and Cameron 2007, Verspoor et al. 2011). According to this theory, language consists of numerous interconnected systems where the whole system can be drastically impacted by small changes in one subsystem or in the initial state of the system (de Bot et al. 2007, van Geert and Dijk 2002). | Soheil Behdarvandirad, 2024
  3. (Example)
     ○ With Complexity Theory, or Complex Dynamic Systems Theory (CDST), as it is called when applied to second language development), the shape that language takes is deferred, always in process, ever-emerging dynamically from interaction. It is helpful to conceive of language as displaying "patterns in the flux," like an eddy in a mountain stream, where the water droplets are continually flowing through while the pattern endures. CDST posits a post-structural ecological view, in which the spatiotemporal context is seen as playing an integral role in affecting development. For this reason, it is counterproductive to separate the learner from the learning process. Individual differences are truly person-oriented, even phenomenological, and mutable. Variability in intra- and inter-individual performance is ubiquitous. Equifinality, autopoiesis, self-organization, interconnectedness, co-adaptation, transdisciplinarity—all are core, foundational features of open complex systems. | Diane Larsen-Freeman, 2019

* COMPLEXONSET

  1. (Optimality Theory) A markedness constraint:
    * COMPLEX ONSET
    Don't produce complex onsets (CCV-).
     | Paul Boersma and Clara Levelt, 2000
  2. (Examples)
     ○ In OT, allowing complex onsets amounts to saying that the constraint militating against complex onsets, namely *COMPLEXONSET, is crucially dominated to the effect that complex onsets surface in the language. This apparent when the constraint interacts with the phonological constraint active in the language.

    * Min-σ ≫ MAXIO ≫ DEPIO ≫ * COMPLEXONSET

    Moroccan Arabic
    Input:
    /bka/ 'cry'
    MAXIO *Min-σ DEPIO *COMPLEXONS
     a. bə.ka         *   
    ☞ b. bka            *
     c. bk   *!         
     d. b.ka      *!      

     Candidate (a) and candidates (c-d) and are ruled out as they violate the higher-ranked constraints DEPIO and MAXIO, respectively. Faithful to the input and incurring no violation to none of the undominated constraints, candidate (b) is selected as the optimal, the surface form. Interestingly, *COMPLEXONSET has no say as to the well-formedness of the selected or the ruled out candidates. As has been noted earlier, this constraint is inactive, as the language allows complex onsets. | Ayoub Loutfi, 2016
     ○ In an underlying word-initial consonant cluster, *COMPLEXONSET is in conflict with the faithfulness constraints MAX-IO (Input segments must have counterparts in the output) and DEP-IO (The output must preserve all segments present in the input). | Somdev Kar, 2012
     ○ This paper focuses on four strategies of onset reduction employed by a single child (4;0-4;4) acquiring Polish: deletion, coalescence, metathesis, and gemination.
     The OT account makes it possible to envisage the four strategies as different surface responses to the undominated *COMPLEXOnset which militates against onset clusters. The choice of a particular strategy as well as its restriction to a particular word position is not random but follows from the interplay between * COMPLEXOnset, sonority-based syllable structure constraints (Margin Hierarchy, CONTACT LAW), context-sensitive markedness constraints (CODA CONDITION, *Nasal-Fricative) and faithfulness constraints. | Beata Łukaszewicz, 2007

Page Last Modified August 30, 2026

 
B a c k   T o   I n d e x