Abstract
This chapter focuses on the interaction of morphological markedness constraints with the feature-deletion and terminal-deletion operations of Impoverishment and Obliteration. We examine both context-free markedness (the marked value of a particular binary feature) and context-sensitive markedness (the marked combination of certain feature-values in the presence of others). A large part of the chapter is devoted to an exemplification of Participant Dissimilation, a process of morphological dissimilation based on multiple instances of the feature [ + participant] in the same M-word. We argue for a general constraint, found across many Biscayan dialects, that bans the co-occurrence of first plural clitics and second person clitics within the same finite verb, but that each dialect may impose additional subcondition and enacts separate repairs in terms of deletion operations of Participant Dissimilation. We exemplify the distinction between Impoverishment and Obliteration through an examination of their effects on the allomorph selection between transitive and intransitive auxiliary roots. This chapter also presents an analysis of the phenomenon of Plural Clitic Impoverishment, whereby the number distinction on absolutive and dative clitics is neutralized in the context of a particular type of ergative clitic. The deletion phenomena in this chapter exemplify some of the procedures recurrently found during the participation of the Feature Markedness module in the Spellout of the Basque auxiliary.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Redundancy rules also may involve insertion of individual features—e.g. the operations that insert default values, such as Halle and Marantz’s (1994) rules of inserting inflectional class features. In fact, our approach to Case (outlined in Chaps. 1–2) treats absolutive as a default case, which we implement in terms of it not being determined in the syntax, but an ‘epenthetic’ postsyntactic feature-insertion (of [ − peripheral, − motion]), in the sense of Trommer (2010).
- 2.
See the following two sections for examples of feature deletion rules.
- 3.
In 3, u stands for the unmarked value of the feature F, and m for the marked value.
- 4.
In Nevins (2007) it was argued that impersonal pronouns are distinct from third person pronouns in that the former remain unspecified for their values of [ ± author, ± participant]. However, it may be the case that what impersonals are missing is not the value for such features, but rather lack these features altogether.
- 5.
Impoverishment of a feature cross-referencing a single argument that is deleted at one site but not another defeats the claim that apparent multiple exponence throughout the Basque auxiliary represents some kind of autosegmental linkage of a single feature to multiple positions.
- 6.
Moreover, the presence of [ + participant] is even more predictable from the presence of [ + author, + singular], since a singular group with only the author must contain exclusively participants, while a [ − singular] group containing the author might include some non-participants. Interesting as this may be, the text demonstrates that this kind of predictability is not relevant to the lifecycle of this morphosyntactic feature, which is present at one point in the derivation and absent thereafter.
- 7.
See Nevins (2007:307–310) for evidence that this is impersonal se, not its reflexive homophone.
- 8.
Spurious se also neutralizes number distinctions, but this is a general property of Spanish se, and not a direct consequence of Impoverishment in clitic clusters.
- 9.
See Tables A.3–A.8 in Appendix A for full paradigms illustrating Participant Dissimilation in Ondarru.
- 10.
Case labels such as ‘ergative’ in this rule and others below are abbreviations for the corresponding case feature sets.
- 11.
The order of application of these two Impoverishment rules is not relevant for the output.
- 12.
See Tables A.3–A.8 in Appendix A for full paradigms illustrating Participant Dissimilation in Zamudio. The effects of Participant Dissimilation are attested in our main source for this variety (Gaminde 2000), as well as in our own field work. Forms missing in these sources have been obtained from de Yrizar 1992b in order to provide full paradigms in Appendix A, especially in the past tense. This accounts for all (apparent) exceptions to Participant Dissimilation in the past tense in this variety found in Appendix A.
- 13.
Note that the ergative clitic in s-endu-e-n in 34 surfaces in proclitic position, due to Ergative Metathesis (Sect. 5.4 in Chap. 5).
- 14.
The L-exponent in this form is y-. This seems to be due to idiolectal variation in the realization of the epenthetic L-morpheme in Zamudio, where many speakers, as reported in Gaminde (2000) and in Tables A.2, A.4 and A.5 in Appendix A, have d- instead in present tense applicative forms of this type. The speaker from whom we obtained 35 seems to have a more general distribution for the L-exponent y- (see Sect. 5.4.3 in Chap. 5 for relevant discussion).
- 15.
Note that the dative clitic undergoes Plural Fission in the Exponence Conversion module (Sect. 3.3.4 in Chap. 3).
- 16.
The auxiliary in 40 is from de Yrizar (1992b), which only provides auxiliary paradigms. We have filled in the rest of the sentence with an overt main verb and subject and object arguments, trying to replicate the context that elicited this auxiliary. de Yrizar (1992b:463) cites two separate sources for his Alboniga data. The first one is field notes by Resurrección María de Azkue, who gathered data from this neighborhood at the beginning of the twentieth century (as part of the field work that formed the basis of de Azkue (1925)). The second is field work conducted by Martín Olazar between 1980 and 1982. The auxiliary form in 40 is from the second source. De Yrizar also provides the form saittuegu, with an overt first plural ergative clitic exponent -gu, obtained from the first source. It seems that Impoverishment of this clitic is a relatively recent innovation in this variety. We have accordingly marked the auxiliary with -gu as ungrammatical in 40.
- 17.
Like Zamudio, all the Basque varieties discussed in this subsection are in the Western Biscayan subdialectal area, according to de Yrizar’s (1992b) classification.
- 18.
On the colloquial/formal distinction in the second person in Basque, see Sect. 1.4.5 in Chap. 1.
- 19.
Other potential Participant Dissimilation forms in Butroi are given in Gaminde (1982:421, 424) and de Yrizar (1992b:636–637). However, it is not clear to us whether the missing overt clitics in these forms are due to phonological processes instead of Impoverishment.
- 20.
In Zamudio and many other Western Biscayan varieties, this has resulted in lexicalization of the first plural clitic as -u (Sect. 3.4.2 in Chap. 3).
- 21.
Although the phenomena discussed in this section are attested in our primary sources for these three varieties (Hualde et al. 1994, Gaminde 2000 and our own fieldwork), we have not undertaken a systematic study of variation with respect to Plural Clitic Impoverishment in Biscayan, so we do not know how widespread it is. Not all sources for these three varieties agree on the data discussed here. For instance, the ditransitive forms given for Lekeitio in de Yrizar (1992b:99) are not subject to Plural Clitic Impoverishment in the context of a participant clitic (see 45 below), and therefore lack the number neutralizations illustrated in Table 4.1. This might be due to idiolectal variation, or perhaps the Impoverishment rule is a relatively recent innovation: de Yrizar’s sources are from early twentieth century and the 1970s and early 1980s, while our source for this dialect was published in 1994. We have found similar discrepancies in the Ondarru and Zamudio data in de Yrizar 1992b. All of them seem to point to idiolectal variation (or historical change) involving absence of one of the Impoverishment rules proposed in this section.
- 22.
We concentrate here on present tense forms. In the past tense, the rules proposed below have a number of exceptions (different ones depending on dialect), and it is not clear to us at this point whether these should be handled as exceptions to the rules, or by positing different Impoverishment rules for the past tense. See Tables A.6–A.8 in Appendix A for the relevant past tense forms.
- 23.
Third person absolutive arguments do not trigger cliticization, and therefore are not relevant for Plural Clitic Impoverishment.
- 24.
This exponent is also present in the context of a first singular ergative clitic. This follows from our analysis, as discussed below.
- 25.
If they originated in the same clitic, this would predict generalized neutralization of number in participant ergative clitics, contrary to fact.
- 26.
Although this instantiation of Plural Clitic Impoverishment does not have an overt effect, it makes a potentially interesting prediction. If in a given variety the entries for first plural clitics are specified for number, Plural Clitic Impoverishment would result in a default realization of the clitic (which would be null or overt, depending on the case and position of the clitic). On the surface, this would have the same effect as Participant Dissimilation targeting a first plural clitic in the context of a second person clitic. Thus, Impoverishment of first plural in the context of a participant clitic would have two sources: Participant Dissimilation and Plural Clitic Impoverishment. This might explain why, as noted in Sect. 4.6, Participant Dissimilation targets first plural more often than second person.
- 27.
‘Mais le basque répugne á répéter l’indice -te, même quand il doit remplir deux fonctions différentes. Ainsi, en labourdin moderne et en guipuzcoan, en regard de dio “il le lui a”, diote peut signifier “ils le lui ont”, “il le leur a”, “ils le leur ont”.’ Lafon (1961:151) Our translation: ‘But Basque is loath to repeat the marker -te, even when it must fulfill two different functions. Thus, in Modern Labourdin and in Guipuscoan, alongside dio “he has it to him”, diote can mean “they have it to him”, “he has it to them”, “they have it to them”.’
- 28.
The auxiliaries in Table 4.1 are shown in their underlying form, so the effect of this phonological rule is not apparent.
- 29.
Recall that absolutive case is not determined in the syntax. See Sect. 1.4.1 in Chap. 1.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer Science+Business Media B.V.
About this chapter
Cite this chapter
Arregi, K., Nevins, A. (2012). Deletion Operations Targeting Morphological Markedness. In: Morphotactics. Studies in Natural Language and Linguistic Theory, vol 86. Springer, Dordrecht. https://doi.org/10.1007/978-94-007-3889-8_4
Download citation
DOI: https://doi.org/10.1007/978-94-007-3889-8_4
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-007-3888-1
Online ISBN: 978-94-007-3889-8
eBook Packages: Humanities, Social Sciences and LawSocial Sciences (R0)