Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  Clause Typing in the Old Irish Verbal Complex

Reviewer: Jean-François R. Mondon
Book Title: Clause Typing in the Old Irish Verbal Complex
Book Author: Carlos García-Castillero
Publisher: De Gruyter Mouton
Linguistic Field(s): Historical Linguistics
Subject Language(s): Irish, Old
Issue Number: 32.1737

Discuss this Review
Help on Posting

This book, a continuation of García-Castillero’s very active research agenda, is a descriptively rich outline of the morphological realizations of clause types in Old Irish. It comprises eleven chapters divided over three parts.

Part I offers the linguistic background necessary for full understanding of clause typing in Old Irish. It consists of three chapters, the first of which begins by summarizing the place of Old Irish in the Celtic family tree and the early glosses (750-850 CE) on three manuscripts: Würzburg, Milan, and St. Gall. The chapter then briefly discusses non-graphic variation observable in the manuscripts, such as the variable position of the perfectivizing morpheme ro- (e.g. contrast niruthógaitsam ‘we have not deceived’ with a form of -ro- immediately following the negative prefix ni- with nimthorgaith ‘it has not deceived me’ with a reflex of -ro- separated from the negative by a lexical prefix associated with the verb). García-Castillero reasonably concludes, in agreement with Russell (2005), that such facts are evidence that the “Glosses represent a rather spontaneous linguistic production, with little or no literary intention on the part of the Old Irish glossators” (p. 15). Chapter 1 continues with a very clear discussion of the morass of Old Irish orthography and it concludes with a presentation of clause typing and its reality as a grammatical category. Clause typing is defined as “the grammatical expression of illocutionary force” (p. 22), which itself “is part of the pragmatic component of the language and corresponds to the intention of the speaker to modify the addressee(s) behavior and/or cognitive situation when the former presents some propositional content” (p. 21). García-Castillero distinguishes six clause types in Old Irish: declarative, relative wh-interrogative, polar interrogative, responsive, and imperative; each is the centerpiece of its own chapter in Part 2.

Chapter 2 is wholly devoted to laying out the blueprint of the Old Irish verbal complex. As opposed to other Indo-European languages whose verbal complex generally consists of just the verbal stem and the inflectional endings, García-Castillero argues that the Old Irish verbal complex additionally contains markers of clause typing. This complex, he maintains, consists of six templatic slots, which to some degree are reminiscent of similar templatic structures proposed for Athabaskan verbs such as Navajo (Hale 2001). Of the six slots, the fourth and fifth are mandatory, consisting of the verbal stem and the inflectional endings respectively. The second slot and the sixth slot are both locations for object affixes, while the first slot houses ‘conjunct particles’ such as in (introduces yes/no questions) or ní (negation). When the first slot is empty, lexical preverbs (e.g. do- in do-beir- ‘give, bring’ from beir- ‘carry, bear’) fall there, otherwise they are relegated to the third slot. The main phonological impetus for treating the verbal complex as a single unit is its being isomorphic with the corresponding stress unit. Stress always falls on the first element after the second slot, be it a lexical preverb in the third slot or absent that, the verbal stem in the fourth slot.

There are both phonological and morphological phenomena which appear to be reflexes of clause typing. The main phonological effect is the mutation on the initial consonant of the stressed element of the verbal complex in relative clauses. This consonant is liable to undergo lenition or nasalization depending upon the gender and number of the antecedent on the one hand, and the grammatical function of the antecedent’s correspondent in the relative clause on the other. As for morphological effects from clause typing, negative clauses furnish such an example. If the sentence is declarative the marker of negation in the verbal complex is ní, if relative it is nad, if relative but followed by an object affix in the second slot then it is nach, if imperative it is na-, and if interrogative the occupant of the first slot is innad. Such variation, which is consistently dependent on clause type, is also observed in the type of pronominal affixes which occur, illuminating the use of what have traditionally been termed by Celticists the Class A, B, and C infixed pronouns.

The third and final chapter of the first part focuses on structures which deviate from the canonical unmarked verb-initial (V1) word order of Old Irish. Two such constructions each involve the placement of a constituent before the main verb. Cleft-sentences place any phrasal constituent before the verb; this constituent itself is preceded by the copula and is itself focused. Left-dislocations, on the hand, allow a noun phrase (NP) to appear sentence initial immediately before the main verb. The NP in such constructions most often indicates the topic of the sentence, though less frequently it too can be used for focus (p. 77 and 79). García-Castillero uses both these constructions in later chapters to account for specific diachronic developments in certain clause types.

Chapter 3 concludes with a rich discussion of another set of non-canonical orders in Old Irish which have received much attention in the literature: tmesis and Bergin’s Law. The former is the separation of the pretonic element of a verbal complex (i.e. the first two slots) from the tonic element. The latter is the appearance of dependent forms of the verb when not in V1 position. Such dependent forms are characterized in simple verbs by a unique set of verbal endings (termed “conjunct endings” as opposed to “absolute endings”) and in compound verbs by the stress falling on the first preverb as opposed to immediately following the first preverb (termed “prototonic” forms as opposed to “deuterotonic” forms). Following Greene (1977) and Kelly (1986), García-Castillero splits up tmesis into four types depending upon the nature of the occupant of the pretonic position of the verbal complex (lexical preverb v. conjunct particle) and when the sentence appears non-clause initially, whether or not the clause is introduced by the copula. García-Castillero follows scholars such as MacCoisdealbha (1976), McCone (2000), and Isaac (2003) in maintaining that these tmesis/Bergin’s Law constructions are artificial constructs which do not represent Archaic Irish syntax, as contra Russell (2005) and Eska (2007, 2008). As such, García-Castillero leaves such constructions out of discussion in the remainder of his study.

Part II consists of four chapters which take the six clause types in turn, offering a formal analysis of each. Chapter 4 targets positive declarative and relative clause types, dissecting the different morphological means which mark both. With respect to inflection, simple verbs (i.e. verbs without a lexical prefix) take the absolute endings in both clause types aside from the 1st pl. and 3rd sg. and pl. which have unique relative desinences in relative clauses. Compound verbs (i.e. verbs with at least one lexical prefix), on the other hand, take the conjunct endings in both clause types. The relative clause type, however, regardless of whether the verb is simple or compound, is differentiated from the positive declarative type by two other means. In relative clauses the onset of the stressed element of the verbal complex is subjected to lenition or nasalization if it is a phonological target for either mutation. Additionally, when an object affix is part of the verbal complex, a unique type of pronominal object infixes, termed Class C, is used in positive relative clauses as opposed to the use of either suffixal pronominal objects or a different type of infixes (Classes A/B) in a positive declarative verbal complex. García-Castillero links the lack of unique relative endings for the non-3rd person (aside from the 1st pl.) to the fact “that the verbal complexes that have a 1st or 2nd person are less prone to appear in restrictive relative clauses” (p. 147). As such, they appear more declarative-clause like. He finds support for this in the occasional appearance of Class A/B object infixes of 1st and 2nd person objects in relative clauses instead of Class C pronouns.

Chapter 5 delves into subordination, delineating the type and context for each of its five main types and subtypes (p. 151). Declarative and relative clause types stand on both ends of the continuum ranging from more to less main-clause-like. Two examples of more main-clause-like subordinate phrases are those with no subordinating conjunction and those with a left-dislocated topic. Since left-dislocation is not possible in relative clauses, it is understandable that relative verbal forms would not be used in such subordinate clauses. Examples of less main-clause-like subordinate clauses are adverbial clauses. Such clauses do take relative morphology; specifically, that of nasalizing relative clauses.

Wh-interrogatives are presented in Chapter 6. García-Castillero methodically goes through all the permutations of the Old Irish wh-elements, including whether the wh-element is stressed or unstressed, whether it is prenominal, preverbal, or pre-pronominal, and whether the wh-element represents a direct case (nominative or accusative) or an oblique one. The declarative verbal clause type is excluded from wh-interrogatives in favor of relative clause typing which itself varies depending on the factors mentioned above. For instance, in preverbal wh-interrogatives, relative verbal forms are employed when the wh-element is stressed while dependent forms alone (i.e. conjunct endings in simple verbs and prototonic forms in compound verbs) with no concomitant relative mutations when the wh-element is not stressed.

Wh-interrogative clauses share with declarative and relative ones their ability to appear in either of the two pragmatically marked constructions: clefts and right-dislocation. Chapter 7 groups together the three remaining clause types which are generally united by their inability to appear in such constructions: polar interrogatives, responsives, and imperatives. García-Castillero maintains that these three types “mainly focus on the verbal predicate itself.” He argues that nominal and pronominal references play a secondary role in all three, being completely excluded from responsives in fact, which ⎯ as the name suggests ⎯ are used to respond to a yes/no-question (Watkins 1963 for the terminology) and also in some type of ‘emotional reflex’ following imperatives (Draak 1952). All three clause types take conjunct endings (aside from the 2nd and 3rd singulars of the imperative which have unique endings), they bear no autonomous mutations not triggered by specific particles, and compound verbs are regularly in the prototonic form, aside from imperatives with an infixed pronoun. Suffixed pronouns are excluded from all three types, those instances which have been claimed to be imperative verbs coupled with an object suffix are better taken as subjunctive forms or corruptions in the text (Breatnach 1977). In accounting for a few unexpected occurrences of what appear to be responsives, García-Castillero makes the interesting proposal that whereas imperative and polar interrogatives are effectively clause types of initiation, as seen in their uses as conditional protases, the responsive is a reactive clause type, which can therefore also be used as the apodosis of a conditional sentence (p. 225-6).

The third and final part of the book consists of three chapters which are effectively devoted to the morphological paradigms of the various clause types. Chapter 8 goes through each clause type in turn, concluding that “the marking of clause typing predominantly appears at the edges of the Old Irish verbal complex” (p. 259) and that polarity and person show the highest degree of interaction with clause typing. Polarity is indicated by negative conjunct particles which occupy slot 1. Person, on the other hand, is indicated via subject inflectional endings in slot 5 and object affixes in either slots 2 or 6. Certain gaps in the paradigms, which he clearly lays out, have natural explanations which go beyond Old Irish. To take but one example, there is no negative wh-interrogative particle or string. Such a combination is cross-linguistically rare (Erteschik-Shir 1992) and another V1 language, the Australian language Wanyi, has a similar restriction (Laughren, Pensalfini, and Mylne 2005).

Chapter 9 spotlights the suppletive distribution of the present forms of the copula and substantive verb according to clause typing. His thorough descriptive treatment is littered with interesting diachronic asides. One such is how precisely ‘fil’, which is etymologically related to Middle Welsh gwelet ‘to see’ (Schumacher 2004: 669-675, Rix et al. 2001: 675), became a relative form. García-Castillero assumes that its development into an expression of locative predicates originated in negative and interrogative clauses. He compares the English sentences ‘You don’t see the headings’ and ‘Do you see (the) headings? Both can mean in the right context ‘There are no headings’ and ‘Are there headings? He maintains that the situation was comparable in Old Irish, which, however, went a step further. On analogy with passive verbal forms whose conjunct forms ⎯ used, for instance, in negative and interrogative sentences (e.g. ní·carthar ‘(s)he is not loved’, in·carthar ‘is (s)he loved’) ⎯ are identical to the positive relative forms (carthar ‘who is loved’), ‘fil’ was likewise extended to positive relative function, which is in fact its most frequent use in the glosses.

Chapter 10 delves into personal pronouns, narrowing in and offering thoughts on the origins of tonic pronouns in referential non-verbal predication and affixal pronouns, bleached of their original meaning, subsuming the role of aspect typing. With respect to the former, a tonic pronoun is required in constructions such as ‘X is Y’ when X is a pronoun. García-Castillero maintains that since the copula is not stressed and since non-verbal predication discloses some sort of unexpected information, it is unsurprising tonic pronouns are used in such predication. He sees a strong parallel with cleft-sentences and wh-questions. What is curious, however, is that outside of the 3rd pl, such tonic pronouns are required in this construction when neither X nor Y are pronouns, resulting in the following literal translation: ‘Is it X Y.’ García-Castillero views this as an incomplete grammaticalization since ‘it’ does agree in gender with one of the nouns in the construction. The second half of the chapter discusses the conditional and reiterative functions of the infix -dL- and traces its emergence from an affixed pronoun. To illustrate the rise of the latter, García-Catillero starts from sentences in which the infix has a clear nominal reference as in Würzburg 9c22, ‘you forgive not the injury that is done to you, but you complain about it’. Examples like this then influenced the use of the infix pronoun in sentences where it effectively referred to a verbal notion mentioned in a preceding clause and not the objects of that verb as in Würzburg 26a20, ‘he will perform false miracles and false signs, as wizards have done it through time’. Such blurring of the antecedent of a pronoun went one step further, in which the pronoun simply came to mark the relationship of a verb to the preceding context. The veracity of this marking is proved by the fact that intransitive verbs, which by their very nature do not have object arguments, came to take the infix -dL- in such a sequential context.

A final short chapter summarizes the various conclusions reached in the book. The book is rounded out with a bibliography of primary and secondary sources and a detailed index.


This book is descriptively rich and offers the specialist or the neophyte alike a deeper understanding and control of the Old Irish data. It effectively could serve as an appendix to Thurneysen (1946), offering a deeper insight into specific constructions. Only a few times does García-Castillero break the mold of description and offer solutions to certain diachronic puzzles. His solutions to each seem reasonable or at least as convincing as the ones he argues against. The book is overwhelmingly descriptive in nature, however, and does not make an attempt at offering a synchronic proposal couched in modern syntactic or morphological theory accounting for the various morphological pieces or processes tied to each clause type, as is done in Arregi and Nevins 2012 description of the Basque auxiliaries. This is in no way a criticism, as this was simply not the book’s intent, and had it been, its 397 pages would have ballooned to something perhaps double in size.


Arregi, Karlos and Andrew Nevins. 2012. Morphotactics: Basque auxiliaries and the structure of spellout. Dordrecht: Springer.

Breatnach, Liam. 1977. “The suffix pronouns in Early Irish. Celtica 12:75-107.

Draak, Maartje. 1952. “Emotional reflexes,” Ériu 16:74-78.

Erteschik-Shir, Nomi. 1992. “Focus structure and predication: the case of negative wh-questions,” Belgian Journal of Linguistics7.1: 35-51.

Eska, Joseph. 2007. ‘Bergin’s Rule (Syntactic diachrony and discourse strategy),’ Diachronica 24:253-278.

Eska, Joseph. 2008. ‘Grammars in Conflict. Phonological aspects of the Bergin’s Rule construction,’ Keltische Forschungen 3: 45-62.

Greene, David. 1977. ‘Archaic Irish.’ In Karl Horst Schmidt and Rolf Ködderitzsch (eds.), Indogermanisch und Keltisch, 11-33. Wiesbaden: Reichert.

Hale, Ken. 2001. ‘Navajo verb stem position and the bipartite structure of the Navajo conjunct sector,’ Linguistic Inquiry 32.4: 678-693.

Isaac, Graham. 2003. ‘Prospects in Old Irish syntax,’ Zeitschrift für celtische Philologie 53: 181-197.

Kelly, Fergus. 1986. ‘Two notes on final-verb constructions,’ Celtica 18:1-12.

Laughren, Mary, Rob Pensalfini, and Tom Mylne. 2005. “Accounting for verb-initial order in an Australian language,” in (A. Carnie, H. Harley, and S. Dooley (eds.), Verb First (On the Syntax of Verb Initial Languages), 367-401. Amsterdam & Philadelphia: Benjamins.

MacCoisdealbha, Padraig. 1976. The Syntax of the Sentence in Old Irish (Selected Studies from a Descriptive, Historical, and Comparative Point of View. Ph.D. dissertation, Ruhr-Universität Bochum.

McCone, Kim (ed). 2000. Echtrae Chonnlai (and the Beginnings of Vernacular Narrative Writing in Ireland: a Critical Edition with Introduction, Notes, Bibliography and Vocabulary). Maynooth: Department of Old and Middle Irish, National Univrsity of Ireland, Maynooth.

Rix, Helmut, Martin Kümmel, Thomas Zehnder, Reiner Lipp, and Brigitte Schirmer. 2001. Lexikon der indogermanischen Verben. Die Wurzeln und ihre Primärstammbildungen2. Wiesbaden: Dr. Ludwig Reichert Verlag.

Russell, Paul. 2005. ‘What was best of every language: the early history of the Irish language.’ In Dáibhí Ó Cróinín (ed.), A New History of Ireland, Volume 1: Prehistoric and Early Ireland, 405-450. Oxford: Oxford University Press.

Schumacher, Stefan. 2004. Die keltischen Primärverben (Ein vergleichendes, etymologisches und morphologisches Lexicon), unter Mitarbeit von Britta Schulze-Thulin und Caroline aan de Wiel. Innsbruck: Innsbrücker Beiträge zur Sprachwissenschaft.

Thurneysen, Rudolf. 1946. A grammar of Old Irish. Revised and enlarged edition. Translated from the German by Daniel A. Binchy and Osborn Bergin. Dublin: Dublin Institute of Advanced Studies.

Watkins, Calvert. 1963. “Preliminaries to a historical and comparative analysis of the syntax of the Old Irish Verb,” Celtica 6: 1-49.
Jean-François Mondon is an Associate Professor of World Languages at Minot State University whose research interests are historical linguistics, morphological theory, and language pedagogy.

Format: Hardback
ISBN-13: 9783110680300
Pages: 397
Prices: U.S. $ 114.99