A Question of Alignment VI: Syntactic Pivot

In this series of blog articles—taken (more or less) straight from the current working draft of chapter 5.4 of the new grammar for better visibility and as a direct update of an old article (“Flicking Switches: Ayeri and the Austronesian Alignment”, 2012-06-27)—I will finally reconsider the way verbs operate with regards to syntactic alignment.

Since we have just dealt with aspects of syntactic alignment in the last installment and found that Ayeri behaves a little oddly with regards to this, it may be interesting to perform another test on declarative statements and their syntactic pivot as well. A simple test which Comrie (1989: 111–114) describes in this regard is to test coreference in coordinated clauses. In coordinated clauses, it seems to be not uncommon for the subject of the second conjunct to drop out. Thus, in English, which behaves very much in terms of NOMACC alignment in this regard, we get the following result:

In the English example in (1), the cat constitutes the coreferential subject in (1d). This NP is the intransitive subject S of (1b) and the agent A of (1a). English thus typically has NOMACC alignment, since it treats S and A alike. In an ERGABS language, then, we would expect the opposite case: S and P should be treated alike. In Dyirbal, we find the situation depicted by the examples in (2).

  1. Dyirbal (adapted from Comrie 1989: 112):

In (2), we find that balan dʸugumbil ‘the woman’ is coreferential in (2d). This is the S of (2c), and the P of (2a). Dyirbal, thus, treats S and P alike, as predicted for an ERGABS language—at least in this case, since Comrie (1989: 113) also explains that 1SG and 2SG pronouns in Dyirbal behave in terms in terms of NOMACC. Comrie (1989) also notes that some languages do not show a clear preference for whether the A or P of the transitive clause in the first conjunct is the preferred reference of the S of the intransitive clause in the second conjunct.

For Tagalog, as Kroeger (1991) explains, “the deletion is not obligatory but null nominative arguments are always interpreted as referring to the nominative argument of the main clause” (30). Due to the way Tagalog treats subjects, however, the nominative argument can be formed by either NP in (3) with the voice marked accordingly on the verb.1

  1. Tagalog (adapted from Kroeger 1991: 31, from Ramoscena 1990: 151–152):

What can be observed in Tagalog is that in (3a), the dropped S argument in the second conjunct, bago umalis … ‘before … leaves’, is coreferential with Marvin, since he is marked as the subject of the first conjunct. Since Marvin is the theme (above marked P for ‘patient’ more generally) of tanong ‘ask’, the clause needs to be marked for objective voice. On the other hand, in (3b), it is Derek who is the subject of the clause, so it is also he who leaves; the verb in the first conjunct clause is marked for actor voice according to the asker as the actor (A) being the subject.

In order to now investigate what the situation is in Ayeri, let us return to our initial set of examples. These examples feature two animals which are treated both as animate neuters. Anaphoric reference is thus potentially ambiguous between paral ‘cat’ and prabara ‘mouse’.

While it is possible in Ayeri to not repeat the coreferential NP in a conjunct clause verbatim, Ayeri still appears to avoid an empty subject slot. Thus, the verb sahayong ‘it comes’ in (4b) displays a pronominal clitic, -yong ‘it’, which constitutes the resumptive subject pronoun of the clause. In (4d) at least, this pronoun is coreferential with the subject in the first conjunct, paral ‘cat’. Seeing as Tagalog switches the subject around by altering the voice marking on the verb, it is certainly illustrative to check how Ayeri fares if the topic is swapped to prabara ‘mouse’.

In (5), the resumptive pronoun is indicated to not refer to the first conjunct’s agent/subject, paral, but to its theme/object, prabara. This may be explained by topicalization: the sentence is about the mouse, so the underspecified argument in the second conjunct, in absence of topic marking that would indicate otherwise, corresponds to the topic. Interestingly, the result is structurally similar to the example of Tagalog in (3) above. It is too early yet, however, to conclude that what was called ‘topic’ so far is the subject; Ayeri is merely not completely unambiguous in this context. Since Tagalog allows any NP of a clause to be the subject, as illustrated by (1) of installment 4 in this series, let us test whether the behavior just described for Ayeri also holds in other contexts of topicalization. The following example presents sentences of differently case-marked topic NPs each, but in every case, the agent NP and the topicalized NP consist of a human referent. Both referents share the same person features so that the verb in the coordinated intransitive clause can theoretically license either of them as its antecedent.

    1. {Yam ilya} {ang Akan} ilonley Maran nay sarayāng.

      yam=il-ya ang=Akan ilon-ley Ø=Maran nay sara=yāng

      DATT=give-3SG.M A=Akan present-P.INAN TOP=Maran and leave=3SG.M.A

      ‘Maran, Akan gives him a present, and he leaves.’ (Maran leaves)

    2. {Na pahya} {ang Maran} ilonley Diyan nay sarayāng.

      na=pah-ya ang=Maran ilon-ley Ø=Diyan nay sara=yāng

      GENT=take.away-3SG.M A=Maran present-P.INAN TOP=Diyan and leave=3SG.M.A

      ‘Diyan, Maran takes the present away from him, and he leaves.’ (Diyan leaves)

    3. {Ya bahaya} {ang Diyan} Maran nay sarayāng.

      ya=baha-ya ang=Diyan Ø=Maran nay sara=yāng

      LOCT=baha-3SG.M A=Diyan TOP=Maran and leave=3SG.M.A

      ‘Maran, Diyan shouts at him, and he leaves.’ (Maran leaves)

    4. {Ri su-sunca} {ang Diyan} ilonley Sedan nay sarayāng.

      ri=su~sunt-ya ang=Diyan ilon-ley Ø=Sedan nay sara=yāng.

      INST=ITER~claim-3SG.M A=Diyan present-P.INAN TOP=Sedan and leave=3SG.M.A

      ‘Sedan, Diyan reclaims the present with his help, and he leaves.’ (Sedan leaves)

    5. {Sā pinyaya} {ang Maran} tatamanyam Sedan nay sarayāng.

      sā=pinya-ya ang=Maran tataman-yam Ø=Sedan nay sara=yāng

      CAUT=ask-3SG.M A=Maran forgiveness-DAT TOP=Sedan and leave=3SG.M.A

      ‘Sedan, he makes Maran ask for forgiveness, and he leaves.’ (Sedan leaves)

In each of the sentences in (6), it is the topicalized NP which is identified as the antecedent for sarayāng ‘he leaves’. Does this mean Ayeri does, in fact, use Austronesian alignment? While the above examples certainly suggest it, let us not forget that the verb in the coordinated clause could theoretically pick either the agent NP or the topicalized NP as its controller. Things look slightly different, however, if the reference of the verb is unambiguous, for instance, because the topicalized argument cannot logically be the agent of the coordinated clause:

In (7), the first conjunct’s verb, as the head of its clause, specifies that the topic of the clause is the patient (P), which is embodied by ilon ‘present’. This NP, however, is not a very typical agent for the verb in the second conjunct, sara- ‘leave’. Besides, this verb is conjugated so as to require an animate masculine controller, whereas ilon is inanimate, as shown by the topic marker le. Ilon is thus not a suitable controller for sarayāng, since their person-feature values clash with each other—the ANIM and GEND values in particular:

    1. ilonN
      (↑ PRED)=‘present’
      (↑ INDEX)=
       (↓ PERS)=3
       (↓ NUM)=SG
       (↓ ANIM)=
       (↓ GEND)=INAN
    2. sarayāngI
      (↑ PRED)=‘leave ‹(↑ SUBJ)›
      (↑ SUBJ)=
       (↓ PRED)=pro
       (↓ PERS)=3
       (↓ NUM)=SG
       (↓ ANIM)=+
       (↓ GEND)=M
       (↓ CASE)=A

As before, there are two masculine NPs in the first conjunct which form suitable antecedents on behalf of being animate masculine as required: the agent (A) Akan and the recipient (R) Maran. Of the remaining non-topic NPs, Ayeri considers the agent to rank higher as a secondary topic on the thematic hierarchy than the recipient. The agent hence forms the preferred controller for sarayāng.

  1. Thematic hierarchy (Bresnan et al. 2016: 329):

    agent > beneficiary > experiencer/goal > instrument > patient/theme > locative

In cases where the topic in the first conjunct can safely be ruled out as the controller of the pronominal in the second conjunct, the syntactic pivot, thus, defaults to the highest-ranking semantically coherent NP. In most cases, Ayeri will therefore group the intransitive subject and the transitive agent together. For most verbs, this is also reflected by case marking, as we have seen above in (4): the S of an intransitive clause receives the same case marker as the A of a transitive clause: -ang/ang for animate referents, and reng/eng for inanimate referents. The case described initially, where the topic marking basically determines the controller of the coordinated intransitive clause, which is reminiscent of Tagalog’s syntax, is essentially a strategy to disambiguate between two possible controllers for the same target.

When only one of the referents in the transitive conjunct is eligible as the controller of the subject of the intransitive conjunct at the same time, A and P are regularly indicated by person agreement, since Ayeri requires a resumptive pronominal clitic in the intransitive clause, as indicated above. The affix on the verb thus has the status of a pronominal predicator, compare (10).

In (10a), the verb in the second conjunct, sarayāng ‘he leaves’ is marked for a masculine third-person subject. The only available controller in the first conjunct is Lita on behalf of being male, since Kumang is female. Hence, in (10b) the verb of the intransitive conjunct, sarayeng ‘she leaves’, finds its controller only in Kumang.

  • Bresnan, Joan et al. Lexical-Functional Syntax. 2nd ed. Chichester: Wiley Blackwell, 2016. Print. Blackwell Textbooks in Linguistics 16.
  • Comrie, Bernard. Language Universals and Linguistic Typology. Syntax and Morphology. 2nd ed. London: Blackwell, 1989. Print.
  • Kroeger, Paul R. Phrase Structure and Grammatical Relations in Tagalog. Diss. Stanford University, 1991. Web. 17 Dec. 2016. ‹http://www.gial.edu/wp-content/uploads/paul_kroeger/PK-thesis-revised-all-chapters-readonly.pdf›.
  • Ramos, Teresita V. and Resty M. Cena. Modern Tagalog: Grammatical Explanations and Exercises for Non-native Speakers. Honolulu, HI: University of Hawaiʻi Press, 1990. Print.
  1. Thus, compare the English passive sentence Marvini was asked by Derekj before hei left with (3a). In English, the reference of he is ambiguous between the syntactic subject Marvin and the agent Derek, however. As we have seen above, though, Tagalog would also be able to make a subject of an oblique argument, not just of the patient/theme or the recipient. The actor of the Tagalog sentence is also basically an object, not demoted to an adverbial as in English (Kroeger 1991: 38–44).

A Question of Alignment V: Verb agreement

In this series of blog articles—taken (more or less) straight from the current working draft of chapter 5.4 of the new grammar for better visibility and as a direct update of an old article (“Flicking Switches: Ayeri and the Austronesian Alignment”, 2012-06-27)—I will finally reconsider the way verbs operate with regards to syntactic alignment.

One of the most prominent features of Ayeri with regards to verbs and their relation to subjects is verb agreement with 3rd-person NPs. This was already discussed at length in two previous blog articles (“Verb Agreement in Ayeri: Bound, Clitic, or Both?”, 2016-06-01; “Clitics in Ayeri: Thoughts and Notes”, 2017-04-16). Hence, I will only give basic information here.

Kroeger (1991) mentions that Tagalog has optional plural agreement of predicates with the nominative NP if the nominative argument of the clause is plural. This is independent of whether the nominative argument is also the actor of the clause or not (Kroeger 1991: 24–25), compare (1). The arrows in (1) mark government and agreement relationships: the verb governs role and case assignment (top arrow), while the nominative NP controls plural agreement on the verb (bottom arrow). As the arrows illustrate, the relationship between the assignment of the subject role and thus nominative case and plural agreement on the verb are congruent: the verb agrees in both (1a) and (1b) with the respective nominative NP, whether it is the agent (1a) or not (1b).

  1. Tagalog (adapted from Kroeger 1991: 14):

As described before, person agreement in Ayeri is essentially fixed to the agent NP in canonical cases, whether it is the topic of the clause or not. In (2a), we can see the verb determine that the agent argument is also the topic, with the verb agreeing itself in person with the agent: Ajān is a male name; the verb corresponds with masculine agreement. In (2b), however, the relation is asymmetric in that the marking on the verb shows that the patient argument is the topic, while the verb still displays masculine person agreement. We know that the verb agrees with Ajān rather than with Pila because the latter is a female name, so the verb should have feminine agreement if it were to agree with the patient NP. However, as the example shows, the verb continues to agree with the agent NP in spite of not being the topic of the clause. Topicalization appears to have no influence on the distribution of person agreement on the verb; the agent NP remains the subject. This is a very NOMACC trait.

In agentless clauses, however, the verb agrees with the patient argument, which makes Ayeri less typical a NOMACC language, and more similar in this regard to what an ERGABS language would be expected to do. Passivization of a transitive clause as a strategy for keeping the topic constant as a subject is essentially preempted by Ayeri’s use of a topic particle in the verb phrase. Hence, a sentence like (3a)—as a parallel to (1b)—sounds odd, while (3b) is fine.

A Question of Alignment IV: Some General Observations

In this series of blog articles—taken (more or less) straight from the current working draft of chapter 5.4 of the new grammar for better visibility and as a direct update of an old article (“Flicking Switches: Ayeri and the Austronesian Alignment”, 2012-06-27)—I will finally reconsider the way verbs operate with regards to syntactic alignment.

As mentioned in a previous article in this series, Ayeri was originally conceived under an impression of what was described in a quotation by Cowan (1995) in terms of ‘trigger language’ (also compare Schachter 2015). That is, in simple declarative statements, the semantic macrorole of a definite NP is marked on the verb. This is itself a very basic account of what can be observed in Tagalog and other Philippine languages, compare (1) below (emphasis mine).1 Further effects—which I completely disregarded for a long time—will be discussed in more detail in the next few blog articles in this series.

  1. Tagalog (Kroeger 1991: 14, adapted from Foley and Van Valin 1984: 135):
    1. B-um-ili ang=lalake ng=isda sa=tindahan.

      PFV.AV-buy NOM=man GEN=fish DAT=store

      The man bought fish at the store.’

    2. B-in-ili-Ø ng=lalake ang=isda sa=tindahan.

      PFV-buy-OV GEN=man NOM=fish DAT=store

      ‘The man bought the fish at the store.’

    3. B-in-ilh-an ng=lalake ng=isda ang=tindahan.

      PFV-buy-DV GEN=man GEN=fish NOM=store

      ‘The man bought fish at the store.’

    4. Ip-in-am-bili ng=lalake ng=isda ang=pera.

      IV-PFV-buy GEN=man GEN=fish NOM=money

      ‘The man bought fish at the store with the money.’

The examples in (1) show variations on the same sentence, differing in the distribution of the definite NP which Kroeger (1991) classifies as being the subject of the respective sentence on syntactic grounds. The subject NPs are marked with the clitic ang, and their role in the clause is reflected by the voice marking on the verb (the root is bili ‘buy’): in (1a) the subject is the actor, in (1b) it is the object, in (1c) it is a location, and in (1d) it is an instrument. What is remarkable is that this voice marking goes beyond mere passivization,2 so even the oblique arguments of (1cd) can become subjects of their respective clauses. Ayeri is at least superficially similar, compare (2).

    1. ang=int-ya ayon-Ø inun-ley moton-ya

      AT=buy-3SG.M man-TOP fish-P.INAN store-LOC

      The man, he bought fish at the store.’

    2. le=int-ya ayon-ang inun-Ø moton-ya

      PT.INAN=buy-3SG.M man-A fish-TOP store-LOC

      The fish, the man bought it at the store.’

    3. ya=int-ya ayon-ang inun-ley moton-Ø

      LOCT=buy-3SG.M man-A fish-P.INAN store-TOP

      The store, the man bought fish there.’

    4. ri=int-ya ayon-ang inun-ley pangis-Ø

      INST=buy-3SG.M man-A fish-P.INAN money-TOP

      The money, the man bought fish with it.’

Like Tagalog, Ayeri marks a privileged NP on the verb, however, in Ayeri, this is the topic, not the subject (this will be subject to further scrutiny later). Unlike in Tagalog, the marked NP is not marked by a particle, but by the very absence of case marking on the NP itself. The marker corresponding to the role of the topic NP appears as a clitic in the shape of the corresponding NP’s case marker in its proclitic form at the left-most edge of the clause, before the verb. While the marker on the verb is thus related to nominal case markers in Ayeri, Tagalog uses a number of affixes for voice marking which are not obviously related to case markers on nouns. For instance, non-subject actors are marked by the genitive clitic ng (pronounced nang), while actor voice is marked by mag- or -um- (Schachter and Otanes 1972: 74, 78; Kroeger 1991: 16–18). In Ayeri, on the other hand, non-topic animate agents are marked on NPs by -ang or ang, and animate agent-topics are marked on the verb by ang as well.

  1. The underlining here is not supposed to be read as marking contrastive focus—this is one of the ‘mistakes’ that has led to what I have in Ayeri, basically, besides then also mixing up focus and topic. It also does not help that terminology is all over the place, as Schachter (2015: 1659) points out.
  2. Note that Kroeger (1991) avoids the terms active voice and passive voice that Schachter (2015) objects to as inappropriate, even though what Tagalog does essentially appears to work along those lines, except in a more generalized way.

A Question of Alignment III: Definition of Terms

In this series of blog articles—taken (more or less) straight from the current working draft of chapter 5.4 of the new grammar for better visibility and as a direct update of an old article (“Flicking Switches: Ayeri and the Austronesian Alignment”, 2012-06-27)—I will finally reconsider the way verbs operate with regards to syntactic alignment.

The terms ‘subject’, ‘topic’, and ‘focus’ were already used a number of times before in this series, but it seems advisable to sketch out working definitions in order to preclude confusion before continuing to look at how Ayeri fares with regards to some of these notions. As we will see, all of subject, topic, and focus relate to different ways in which the relative prominence of certain NPs is raised; subject and topic are also closely related to each other. It ought to be noted that while LFG treats topic and focus as grammaticalized discourse functions outside of the argument-structure frame of a verb, it treats the subject as both a discourse function and an argument function; topic and focus, on the other hand, must be identified with a corresponding argument function, for instance, SUBJ or OBJ (Bresnan et al. 2016: 99–100).


First things first, the subject can be defined in a variety of ways, and maybe especially because the notion of a subject is so basic, Comrie (1989) notes that if

linguists were invariably in agreement in stating which noun phrase, in each construction in each language, is the subject, then we could, perhaps, accept this inter-subjective agreement, and devote correspondingly less energy to trying to find an explicit definition of subject. However, it turns out that, in a wide range of cases, this inter-subjective agreement is lacking. (Comrie 1989: 104)

Dixon (2010) defines a subject as “the entity about which something is affirmed or denied” (76). He goes on to explain that, ignoring copula clauses like ‘We are tired and thirsty’, every language has two varieties of clauses, intransitive ones, where the verb has just one core argument, and transitive ones, where the verb has two core arguments. A basic definition based on this is given by the chart in (1).

    1. nominative–accusative alignment (S/A—P):

    2. ergative–absolutive alignment (S/P—A):

The chart in (1) shows the definition of the notion of subject for both nominative–accusative languages and ergative–absolutive languages. Languages of the world differ based on how they prefer to treat the two nominal relations of a transitive verb in relation to intransitive verbs: they may have a strong preference to either treat the agent (A)—the entity that prototypically acts in some way—or the patient/undergoer/theme (P)—the entity which is prototypically affected by the action in some way—the same as S, the sole argument of an intransitive verb. In the former case, the language is said to have NOMACC alignment (1a) (S/A is the ‘nominative’ subject), whereas in the latter case, the language is said to have ERGABS alignment (1b) (S/P is the ‘absolutive’ subject). Comrie (1989) illustrates this difference with an example from Chukchi, which we will here contrast with English:1

  1. Chukchi (adapted from Comrie 1989: 104):

While English treats the actor of the intransitive sentence (2a) the same as that of the transitive one (2b)—both sentences use I in the nominative—Chukchi appears to use a different pronoun for the actor of the intransitive sentence (3a) than the actor of the transitive one (3b)—absolutive ɣəm versus ergative ɣəmnan, respectively. At least in Standard English, it would be ungrammatical to use the pronoun me in place of I in (2b), since me can only be used for first-person objects of the verb, but not for subjects of transitive clauses.

However, Comrie (1989) also urges to consider that grammatical relations and their representation in morphology are not always as clear-cut as in the example above. While he characterizes the prototypical subject as the intersection of agent and topic as far as cross-linguistic evidence is concerned (107), he also points out that subjects do not necessarily have to unite all the properties typically associated with them (110). This seems to be the case with Tagalog, for instance, as observed by both Schachter (1976) and Kroeger (1991), and may considerably complicate making a definitive statement.

Moreover, Comrie (1989) points out that statistically, languages of the world show a strong preference for NOMACC alignment, possibly due to the fact that human perception values actors as more relevant to discourse than patients, which is why actors are far more likely also to be pragmatic topics (120). Yet, though, dominantly NOMACC-aligned languages may show a bias towards an ERGABS treatment, for instance, of resultative constructions. On the other hand, dominantly ERGABS languages show a bias towards a NOMACC treatment, for instance, of addressees of imperatives (116–119).

According to Carnie (2013), from the point of view of constituent structure (which is key in Generative Grammar), a subject is conventionally understood as a “DP that has the property indicated by the predicate phrase. What the sentence is about. In most sentences, this surfaces in the specifier of [the tense phrase]” (221). However, as we have seen above, this notion is challenged by languages such as Tagalog (Kroeger 1991: 225). What Carnie (2013) refers to in terms of constituent structure is basically indicated by (4). For systemic reasons, Carnie (2013) refers to a DP subject which serves as the specifier of a TP. This corresponds to the subject NP and the IP here. Unlike GG, LFG treats tense as a semantic feature, not as a functional head with a fixed position in constituent structure, hence the difference in labeling.

LFG defines a subject function, SUBJ. Which argument of the verb the subject is mapped onto is understood to be based on the relative prominence of the subject argument along some dimension compared to other arguments. For instance, NOMACC languages prefer the semantically most prominent available role of a verb’s argument structure, ERGABS languages instead pick the argument most affected by the actor’s action, and active languages focus on the argument in control of the action (Bresnan et al. 2016: 95–96). The mapping between grammatical functions like SUBJ and the lexical components that make it up also does not need to be a one-to-one correspondence, since LFG allows for the distributed exponence of grammatical features like in the example of Warlpiri in (5). The only condition is that grammatical functions be uniquely defined within their minimal f-structure (Bresnan et al. 2016: 45). As (5) shows, multiple NPs in different positions in the constituent structure may feed semantic information to a single function defined by the argument structure of the verb.

  1. Warlpiri (Bresnan et al. 2016: 325):

The subject role θ̂ is defined at least in the context of English as “the most prominent semantic role of a predicator” (Bresnan et al. 2016: 330). Furthermore, Bresnan et al. (2016) devise two a-structure features, [± o] (objective) and [± r] (restrictive). According to this classification, SUBJ is assigned the features [– r, – o], since the subject is not restricted to a certain semantic role, nor needs to have a semantic role.2 Also, subjects do not complement transitive predicators like objects do, so they are not ‘objective’. Bresnan et al. (2016)’s lexical mapping theory assumes that all languages have subjects, which goes counter to Schachter (1976, 2015)’s claim that subjects are possibly not universal (Bresnan et al. 2016: 330–331).


The notion of topic refers essentially to who or what a longer stretch of conversation is about. Givón (1983) defines the topic of a ‘thematic paragraph’—as he calls a coherent unit of discourse above the level of a single sentence—as “the continuity marker, the leitmotif” (8). The topic is thus

the participant most crucially involved in the action sequence running through the paragraph; it is the participant most closely associated with the higher-level “theme” of the paragraph; and finally, it is the participant most likely to be coded as the “primary topic”—or grammatical subject—of the vast majority of sequentially-ordered clauses/sentences comprising the thematic paragraph. (8)

This indicates that topic and subject are closely related concepts, as already mentioned above in reference to Comrie (1989). Languages employ various means to indicate topics; right- and left-dislocation, as known from English, or topic-marking particles as in Japanese and Korean, are only two among many possibilities (Dixon 2010: 174).

Topicality also interfaces with definiteness in that chain-initial topics may be definite (already introduced into discourse) or indefinite (newly introduced into discourse), while chain-medial topics and chain-final topics are always expected to be definite (Givón 1983: 10). Dixon (2010: 171) adds that topic NPs are coreferential with arguments of clauses immediately preceding or following the current clause. Moreover, the strategy of passivization (in NOMACC languages) or of antipassivization (in ERGABS languages) exists, among others, in order to keep a certain discourse item persistent in the highly topical subject position even if it would otherwise be the object of the clause. This is related in turn to the notion of syntactic pivot in clause coordination (172).


Regarding the definition of focus, Dixon (2010: 174) only mentions contrastive focus, which basically raises the prominence of a certain NP within a single clause. It is not necessary for the focussed NP to be coordinated with another NP by ‘or’. Dixon (2010) also warns that focus is often confused with topic. Perhaps this is in part also, as Bresnan et al. (2016) mention, due to the fact that English may use the topic position for either topic or focus under certain circumstances (98):

  1. Q: What did you name your cat?
    A: ROSIE I named her. (Rosie = FOC)

The answer to a wh-question is considered focused, so Rosie in (6) is the focus in ‘I named her ROSIE’. However, in the example above, Rosie is fronted, which following Givón (1983), constitutes a disruptive action used to establish a new topic of conversation: left-dislocation in languages with rigid SVO word order such as English is typically associated with low topic continuity, and left-dislocated NPs can be found most often as initiating a topic chain (32).

  • Bresnan, Joan et al. Lexical-Functional Syntax. 2nd ed. Chichester: Wiley Blackwell, 2016. Print. Blackwell Textbooks in Linguistics 16.
  • Carnie, Andrew. Syntax: A Generative Introduction. 3rd ed. Chichester: Wiley Blackwell, 2013. Print. Introducing Linguistics 4.
  • Comrie, Bernard. Language Universals and Linguistic Typology. Syntax and Morphology. 2nd ed. London: Blackwell, 1989. Print.
  • Dixon, Robert M. W. Methodology. Oxford: Oxford UP, 2010. Print. Vol. 1 of Basic Linguistic Theory by Robert M. W. Dixon. 3 vols. Oxford: Oxford UP, 2010–12.
  • Givón, Talmy. “Topic Continuity in Discourse: An Introduction.” Topic Continuity in Discourse: A Quantitative Cross-Language Study. Ed. Talmy Givón. Amsterdam: John Benjamins, 1983. 1–41. Print. Typological Studies in Language 3.
  • Kroeger, Paul R. Phrase Structure and Grammatical Relations in Tagalog. Diss. Stanford University, 1991. Web. 17 Dec. 2016. ‹http://www.gial.edu/wp-content/uploads/paul_kroeger/PK-thesis-revised-all-chapters-readonly.pdf›.
  • Schachter, Paul. “The Subject in Philippine Languages: Topic, Actor, Actor-Topic, or None of the Above?” Subject and Topic. Ed. Charles N. Li. New York: Academic P, 1976. 493–518. Print.
  1. In English, you is the same for both singular and plural as well as subjective and objective case, which is why I replaced it with the unambiguous her in (2).
  2. This ought to make Kroeger (1991)’s analysis compatible to LFG as well.

A Question of Alignment II: ‘Trigger Languages’

In this series of blog articles—taken (more or less) straight from the current working draft of chapter 5.4 of the new grammar for better visibility and as a direct update of an old article (“Flicking Switches: Ayeri and the Austronesian Alignment”, 2012-06-27)—I will finally reconsider the way verbs operate with regards to syntactic alignment.

The notorious term ‘trigger language’ comes up in discussions on Conlang-L as early as 1995, where it may well have originated as an established term in the fictional-language community for what will be described below in brief. That is, I have not been able to find any earlier mentions of the term ‘trigger’ as referring to an alignment system in the archives; other mainstays of the fictional-language community, such as the ZBB, were established only about a decade later. In a message dated December 16, 1995, John Cowan writes that he wants “to propose a reform of Radilu, to make it use the Tagalog concept of a trigger” (Cowan 1995). By his definition, this entails that

each clause contains one noun phrase which is not marked for case, but rather has a distinct marking called the “trigger marker”. […] The verb carries a marking (which of course looks nothing like the noun case markers) that tells the true case of the trigger. […] This involves changing the name of “nominative” and “accusative” to “actor” and “patent” [sic], since there is no longer a “subject” or “object” as such. Of course, word order is free (Cowan 1995)

He also notes that “Usually the trigger is definite (Tagalog doesn’t have articles)” (Cowan 1995). Essentially, it seems that the motivation for Cowan’s system is that the ‘trigger’ indicates that a certain NP is definite. As we will see further on, this is similar to how Tagalog marks one of its relations on the verb, with that relation being definite. Things are more complicated in reality, though. Especially the claim that Tagalog lacks subjects and objects is problematic. However, the term ‘trigger’ seems to have currency in that, for instance, Schachter (2015) chooses it explicitly to refer to the “non-case-marked argument” (1659). In a parenthetical remark he adds that some

previous treatments have referred to the argument in question as the topic and some as the subject. However, as will become clear below, each of these labels appears to carry some inappropriate connotation, making a netural term like Trigger seem preferable […] There also seems to be good reason to reject the term focus. (Schachter 2015: 1659)

It may be noted that term ‘focus’ is used in Schachter and Otanes (1972), the main reference grammar of Tagalog. What is interesting in comparing Schachter (2015)’s and Kroeger (1991)’s respective analyses of Tagalog’s syntactic alignment is that both make the same observation in spite of coming to opposite conclusions: Tagalog is ambiguous as to whether the subject notion is vested in the NP whose role is marked on the verb or the actor, since certain syntactic constructions typically associated with subjects apply to either or both. While this ambiguity leads Schachter (1976, 2015) to ultimately conclude that Tagalog lacks a single unified relation which can be analyzed as a syntactic subject,1 Kroeger (1991) reaches the opposite conclusion by performing further tests and taking a functionalist rather than purely structuralist perspective. Thus, he concludes:

  • “Tagalog has a well-defined grammatical subject” (225). What Schachter (1976) lists as evidence against are special cases which can be explained by the high semantic and pragmatic prominence of actors more generally (Kroeger 1991: 225). Tagalog basically applies the the notion of a logical subject distinct from the syntactic subject to some constructions, though the syntactic subject is more important overall (36).
  • “grammatical relations are defined independently of phrase structure” (225);2
  • “patients can become subjects even when the agent is expressed as a direct (non-oblique) argument of the verb” (225).
  • “Subject selection in Tagalog does not work by demotion or suppression of thematically more prominent arguments. Rather, all arguments seem to be equally eligible for mapping onto the subject relation” (226).

Kroeger (1991) also provides evidence based on statistics and examples that the marked-for relation, which he classifies as being in the nominative case according to his hypothesis that it is the syntactic subject, is neither especially salient in terms of pragmatic topichood, nor does it show signs of carrying pragmatic focus specifically. He finds that rather, nominative marking works independent of discourse functions (56 ff.). All things considered, the term ‘trigger language’ is probably ill-fitting, not just for Ayeri.

The tests for typical properties associated with grammatical subjects which Kroeger (1991) performs partially extend those presented in Schachter (1976). Moreover, his conclusions build on a more modern, functionally oriented approach than Schachter’s. For this reason, I will follow Kroeger rather than Schachter. Either way, in order to compare what is going on in Ayeri, we will have to test verb agreement, syntactic pivot, relativization, control of secondary predicates, raising, and control.3 First of all, it will be helpful, however, to define some terms which will be used in the discussion further on.

  1. Cowan (1995)’s sketch may be based on Schachter (1976). Curiously, Schachter (2015) does not acknowledge Kroeger (1991) at all, nor does he refer to any other research more recent than 1985. The reason may be that Schachter retired in the early 1990s, as the UCLA linguistics department’s Department history suggests.
  2. This point especially may be a problem for generative theories of syntax.
  3. The tests which Kroeger (1991) dismisses as irrelevant to determining subjecthood in Tagalog have been omitted here if they were also not profitable to answering this question for Ayeri. The same goes for a number of tests which are specific to the grammar of Tagalog and thus have no application in Ayeri.

A Question of Alignment I: Typological Considerations

In this series of blog articles—taken (more or less) straight from the current working draft of chapter 5.4 of the new grammar for better visibility and as a direct update of an old article (“Flicking Switches: Ayeri and the Austronesian Alignment”, 2012-06-27)—I will finally reconsider the way verbs operate with regards to syntactic alignment.

Verbs govern the relations of the various phrase types to each other and they are thus central to the formation of clauses. Just from looking at the numerous examples given both on this website and in the grammar, it should be clear that Ayeri’s preferred word order is verb-first, which opens up a few typological questions—first and foremost, whether Ayeri actually has a verb phrase, or in terms of generative grammar: whether it is configurational in this regard. Ayeri definitely has a constituent structure as far as NPs, APs, PPs, etc. are concerned. However, due to VSO word order, it is not obvious whether verb and object actually form a VP constituent together, since V and O are not adjacent to each other. Since Ayeri marks topics in terms of morphology, it will also be necessary to discuss how this mechanism works and how it relates to the notion of the subject.

A discussion of subject, topic, and configurationality is interesting also in that Ayeri’s syntactic alignment was originally inspired by the Austronesian or Philippine alignment system, though then under the term ‘trigger language’ which is itself not unproblematic. Tagalog, an Austronesian language of the Malayo-Polynesian branch, spoken mainly in the Philippines (Hammarström et al. 2017: Tagalog; Schachter and Otanes 1972), usually serves as the academic poster child in descriptions of Austronesian alignment. Ayeri departs from Tagalog’s system in a number of ways, though, and probably towards the more conventional. Austronesian alignment is not necessarily the best model to liken Ayeri’s syntax to. It will nonetheless be informative to compare both systems based on the work of Kroeger (1991, 1993), who provides an analysis of Tagalog’s syntactic alignment roughly in terms of the LFG framework and describes some heuristics which may be helpful in establishing what is actually going on in Ayeri. As mentioned in a previous blog article (“Happy 10th Anniversary, Ayeri”, 2013-12-01), I started Ayeri in late 2003—then still in high school and not knowing much about linguistics. Of course, I had to go and pick as a model the one alignment system which has long been “a notorious problem for both descriptive grammarians and theoretical syntacticians” to the point where it “sometimes seems as if Austronesian specialists can talk (and write) of nothing else” (Kroeger 2007: 41).

As mentioned above, Ayeri’s unmarked word order gives the verb first, and then, in decreasing order of bondedness to the verb, the phrases which make up the verb’s arguments: subject (agent), direct object (patient), indirect object (dative), followed by adverbials in the genitive, locative, instrumental, and causative case. Ayeri’s basic word order is thus VSO, a trait it has in common with about 7 % of the world’s natural languages according to Dryer (2013). Regarding word order typology, we can declare the generalization in (1), which is consistent also with word order in other areas of the language, where the head precedes the modifier. The head is here represented by the verb, the modifier by the object—like English, Ayeri is a VO language, thus. In addition to this, however, Ayeri regularly puts the verb as the head of the clause itself first.

    1. Order of subject, object and verb: VSO
    2. Order of verb and object: VO

It is commonly assumed that languages have a subject which occupies a certain position in the constituent structure—the predicate—and which commands a constituent jointly formed by the verb and its dependents—the predication. An SVO sentence in English thus very generally looks like in (2) (compare the examples in Bresnan et al. 2016: 101–111).

However, Ayeri is a VSO language, so the question arises how the basic constituent structure should be diagrammed in tree form, since V and O are not adjacent. As an initial hypothesis one might assume that they cannot form a unit together, since S somehow stands in between the constituents it is supposed to command. A very first stab at diagramming would probably be to come up with a flat, non-configurational structure, all but lacking a VP, as shown in (3).

  1. ?

Such a structure, though, does not do Ayeri justice in that, for instance, right-node-raising of a subject and object NP together is possible, so there is evidence that they form a constituent subordinate to the verb. NP–XP constructions where XP is not a maximal projection of a verb also exist in isolation, so NP and XP are probably contained in a small-clause constituent S separate from the verb. The verb in the initial position furthermore shows inflection, so one might rather construe it as an I⁰, projecting an IP, which frees up VP for other purposes while we can use IP to govern both Iʹ and S. In fact, such a structure is basically the conclusion Chung and McCloskey (1987) come to for Irish, which is also a VSO language (4a). Bresnan et al. (2016) give the chart in (4b) for Welsh, equally a VSO language (also compare Dalrymple 2001: 66, sourcing Sadler 1997). Kroeger (1991) suggests the two structures depicted in (4c) for Tagalog, based on the suggested constituent structure for Celtic languages.

    1. Irish (Chung and McCloskey 1987: 235):

    2. Welsh (adapted from Bresnan et al. 2016: 134):

    3. Tagalog (Kroeger 1991: 131):

What all of these c-structures have in common is that the inflected verb appears in I⁰, which is a sister of S. S, in turn, is a small clause containing the arguments of the verb. In the cases of Irish and Welsh, however, there is a VP sister of the subject NP which itself does not have a head, but contains the object NP as a complement. In the case of Tagalog, S is non-configurational, that is, while XP may contain a non-finite verb, the subject and object NPs are on equal footing.

Bresnan et al. (2016: 129–138) inform that the phenomenon of the verb ending up in a different head position (V⁰ apparently moves to I⁰) in (4b) is commonly known as ‘head movement’, except that LFG is built specifically without any movement. Since LFG is based on the assumption that all nodes in a syntactic structure are base-generated, that is, that there are no transformational rules generating the surface structure from a deeper layer of representation underneath it, there cannot be a trace of V left behind in VP. LFG avoids empty categories, as there is no information contained in an empty node. The functional information provided by the verb is not lost, however, it is merely now provided by the verb in I⁰. Essentially, the Welsh example does not violate endocentricity, since the finite verb in I⁰ still forms the verbal head in the functional structure representation of the clause. With regards to constituent structure, V⁰, if present, c-commands its NP sister; both V⁰ and NP are dominated by VP:

    1. Exhaustive domination (Carnie 2013: 121):

      “Node A exhaustively dominates a set of terminal nodes {B, C, …, D}, provided it dominates all the members of the set so that there is no member of the set that is not dominated by A and there is no terminal node G dominated by A that is not a member of the set.”

    2. C-command (Carnie 2013: 127):

      “Node A c-commands node B if every node dominating A also dominates B, and neither A nor B dominates the other.”

The AVM in (4b) shows that the contents normally found in V⁰ are provided by the head of its equivalent functional category, I⁰. I⁰ and VP are said to map into the same f-structure (Bresnan et al. 2016: 136). Endocentricity still holds in that IP dominates all nodes below it, thus also I⁰ and the object NP. In addition, I⁰ c-commands its sister node and all of its children, hence also the object NP. As Bresnan et al. (2016) put it: “X is an extended head of Y if X is the Xʹ categorial head of Y […], or if Y lacks a categorial head but X is the closest element higher up in the tree that functions like the f-structure head of Y” (136). For our example, replace X with I⁰ and Y with VP in the second half of the quote: I⁰ is the closest element higher up in the tree that functions like the f-structure head of VP, which itself lacks a categorial head.

The analysis of the sentence structure of Celtic languages shows that VSO languages do not automatically need to be considered ‘non-configurational’ and lacking a VP if the notion of extended heads is accepted. In any case, tests need to be performed to see whether one of the analyses presented in (4) holds true for Ayeri as well. However, this will not be in the scope of this series of blog articles.

  • Bresnan, Joan et al. Lexical-Functional Syntax. 2nd ed. Chichester: Wiley Blackwell, 2016. Print. Blackwell Textbooks in Linguistics 16.
  • Carnie, Andrew. Syntax: A Generative Introduction. 3rd ed. Chichester: Wiley Blackwell, 2013. Print. Introducing Linguistics 4.
  • Chung, Sandra, and James McCloskey. “Government, Barriers, and Small Clauses in Modern Irish.” Linguistic Inquiry 18.2 (1987): 173–237. Web. 11 Aug. 2017. ‹http://www.jstor.org/stable/4178536›.
  • Dalrymple, Mary. Lexical Functional Grammar. San Diego, CA: Academic Press, 2001. Print. Syntax and Semantics 34.
  • Dryer, Matthew S. “Order of Subject, Object and Verb.” The World Atlas of Language Structures Online. Eds. Matthew S. Dryer and Martin Haspelmath. 2013. Max Planck Institute for Evolutionary Anthropology, n.d. Web. 11 Aug. 2017. ‹http://wals.info/chapter/81›.
  • Hammarström, Harald et al., eds. “Language: Tagalog.” Glottolog. Version 3.0. Max Planck Institute for the Science of Human History, n.d. Web. 11 Aug. 2017. ‹http://glottolog.org/resource/languoid/id/taga1270›.
  • Kroeger, Paul R. Phrase Structure and Grammatical Relations in Tagalog. Diss. Stanford University, 1991. Web. 17 Dec. 2016. ‹http://www.gial.edu/wp-content/uploads/paul_kroeger/PK-thesis-revised-all-chapters-readonly.pdf›.
  • ———. “Another Look at Subjecthood in Tagalog.” Pre-publication draft. Philippine Journal of Linguistics 24.2 (1993): 1–16. Web. ‹http://www.gial.edu/documents/Kroeger-Subj-PJL.pdf
  • ———. “McKaughan’s Analysis of Philippine Voice.” Piakandatu ami Dr. Howard P. McKaughan, 41–. Eds. Loren Billings and Nelleke Goudswaard. Manila: Linguistic Society of the Philippines and SIL Philippines, 2007. Print.
  • Sadler, Louisa. “Clitics and the Structure-Function Mapping.” Proceedings of the LFG ’97 Conference, University of California, San Diego, CA. Eds. Miriam Butt and Tracy Holloway King. Stanford, CA: CSLI Publications, 1997. Web. 12 Aug. 2017. ‹https://web.stanford.edu/group/cslipublications/cslipublications/LFG/2/lfg97sadler.pdf›.
  • Schachter, Paul and Fe T. Otanes. 1972. Tagalog Reference Grammar. Berkeley: U of California P, 1983. Google Books. Google, 2011. Web. 6 Nov. 2011. ‹http://books.google.com/books?id=E8tApLUNy94C›.

Update on the Grammar Writing Process IV

Grammar writing has gone slowly again for the past couple of weeks, which is mostly due to reading up on things. I have now arrived at discussing verbs, which are the most complex part of speech since they are at the head of clauses—not just structurally, but also functionally. Important questions right now are:

  • What evidence is there for a constituent S which holds all the verb’s arguments besides the fact that verbless clauses exist complete with predication?
  • Is there a VP in hiding? This requires performing tests on constituency as well (there is a way to say does so as well, so there should be a VP even if the verb word itself is the head of the superordinate IP).

This is to say, I assume that Ayeri’s basic sentence structure looks essentially like this:

The sentence 'Ang konja Yan pahiley' ('Yan eats a cookie') charted in terms of LFG

And then, there are some further questions which I’d like to answer:

  • Austronesian alignment gave the impetus for Ayeri’s strategy of marking one certain NP on the verb, however, after reading Kroeger (1991) it became clear to me that there are strong differences between the real thing and what I have. This is mostly due to not consistently following the original model but falling back on structures familiar from German and English. Thus: what is a so-called ‘trigger conlang’ of which Ayeri is supposedly a prominent example,1 and how is Ayeri actually positioned in this regard?
  • In consequence, how does Ayeri deal with more complex sentence structures, for instance, involving raising and control, as opposed to what Kroeger (1991) describes?
  • Ayeri basically grammaticalizes topic marking by way of agreement morphology. How (un)typical is this with regards to typology? (e.g., see Li and Thompson 1976 for something very old and basic)
  • Does the way in which Ayeri deals with topicalization have any effects on binding? Topics are supposed to operate outside of the functional hierarchy which Bresnan et al. (2016) propose as an important factor in pronominal binding.
  • Since I’ve been trying my hands on an LFG-based analysis, how do verbs behave regarding assigning roles in argument structure? (Dalrymple 2001: 203–215, Bresnan et al. 2016: 329–348)

To be honest, when I started working on Ayeri in 2003, I would not have understood a word of what Kroeger (1991) writes, so it was basically clear from the beginning that there’d be large inconsistencies with regards to the intention of playing around with Austronesian alignment. The thing is, besides Tagalog’s infamous marking of the ang phrase’s role on the verb (actor, goal, direction, beneficiary, etc.), whatever that phrase is syntactically, It also has effects on raising, control, and binding, which I have long ignored out of a lack of knowledge and awareness of these grammatical processes. Even when I tried to come to terms with Ayeri’s syntactic alignment in an often-clicked blog article in 2012, I applied some of the tests discussed there only mechanically, without actually understanding what they’re about.

It also may be noted that Kroeger (1991) analyzes It as the subject because of consistencies with syntactic traits usually associated with subjects, though with the added complication that it’s not fixed to its conventional position as the specifier of VP.2 You can also see It variously analyzed as focus or topic, which is terribly confusing especially when you don’t know a lot, and this confusion had a major impact on what I ended up with in Ayeri. It will also be necessary, thus, to look at whether the logical subject and the syntactic subject in Ayeri coincide. My gut feeling is that they do, which would make Ayeri more similar, in fact, to analyses of the basic clause structure of Celtic languages such as Welsh or Irish (compare, for instance, Chung and McCloskey 1987, Sadler 1997, Dalrymple 2001: 66, Bresnan et al. 2016: 130–138).

  • Bresnan, Joan et al. Lexical-Functional Syntax. 2nd ed. Chichester: Wiley Blackwell, 2016. Print. Blackwell Textbooks in Linguistics 16.
  • Chung, Sandra, and James McCloskey. “Government, Barriers, and Small Clauses in Modern Irish.” Linguistic Inquiry 18.2 (1987): 173–237. Web. 11 Aug. 2017. ‹http://www.jstor.org/stable/4178536›.
  • Dalrymple, Mary. Lexical Functional Grammar. San Diego, CA: Academic Press, 2001. Print. Syntax and Semantics 34.
  • Kroeger, Paul R. Phrase Structure and Grammatical Relations in Tagalog. Diss. Stanford University, 1991. Web. 17 Dec. 2016. ‹http://www.gial.edu/wp-content/uploads/paul_kroeger/PK-thesis-revised-all-chapters-readonly.pdf›.
  • Li, Charles N. and Sandra A. Thompson. “Subject and Topic: A New Typology of Language.” Subject and Topic. Ed. Charles N. Li. New York: Academic P, 1976. 457–485. Print.
  • Sadler, Louisa. “Clitics and the Structure-Function Mapping.” Proceedings of the LFG ’97 Conference, University of California, San Diego, CA. Ed. by Miriam Butt and Tracy Holloway King. Stanford, CA: CSLI Publications, 1997. Web. 12 Aug. 2017. ‹https://web.stanford.edu/group/cslipublications/cslipublications/LFG/2/lfg97sadler.pdf›.
  1. The oldest message on Conlang-L (itself the oldest conlanging group on the internet I’m aware of) which uses the term ‘trigger’ to refer to case/voice marking I could find is by John Cowan, dated December 16, 1995. The archives 1991–1997 seem to only survive archived by the Wayback Machine anymore. Search for the time stamp, “Sat Dec 16 13:09:06 1995”, on the linked archive page to read the message.
  2. This is probably not much of a problem for the likes of LFG or HPSG, but likely more of a problem for generative grammar.

Update on the Grammar Writing Process III

I’ve recently done a lot of proofreading of basically anything besides the introduction chapter of the new Ayeri Grammar. I did this to weed out errors I’ve previously overlooked and also to make sure that what I’d written earlier in the morphology chapter was consistent with the rather extensive work I did in order to come to terms with why certain pre- and suffixes should be clitics. This detour took quite a while—from January to April—but it was probably worth it, since it clarified some questions I had. My quest for clarity on clitics versus affixes in Ayeri culminated in a lengthy blog article, a version of which, revised in parts, can be found in the new grammar as section 3.2.5.

Starting to document Ayeri’s syntax is the logical next step now after I tried to describe its phonology and morphology as well as I could. So, what I’m up to now is trying to describe the morphosyntactic structure of the various syntactic constituents: noun phrases, verb phrases, adjective phrases etc. Since there is very little agreement morphology in Ayeri, this should be rather straightforward for the most part, except for verb phrases (I recently discovered that Dalrymple (2001) contains a whole lot more examples than Bresnan (2016), so that might come in handy). Currently, however, I’m still only concerned with NPs and DPs. What’s still making me self-conscious about all this is that I still have never really studied syntax formally, as I pointed out earlier. So, if you take a look at the grammar and see something implausible, please let me know!

When I tried to figure out clitics in Ayeri earlier, I also came up with a lot of examples of coordination, and one thing I wondered is if the following is actually reasonable.

An attempt to describe formally the distribution of the progressive clitic over two coordinated verbs

What you can see here is an attempt to apply LFG to an example sentence which contains a coordinated constituent: manga sahaya rangya ‘is coming home’ is coordinated with nedraya ‘sits (down)’. The question now is, how to formally describe that manga as the (enclitic) progressive marker is to be understood as distributing over both verbs, sahaya ‘comes’ and nedraya ‘sits’? I actually looked up a few articles (Belyaev et al. 2015; Kaplan and Maxwell 1988; Maxwell and Manning 1996; Peterson 2004) and at least took a casual glance at them, but nowhere did I see any discussion of how to indicate when certain markers in the verb phrase distribute to multiple conjuncts. Instead, I could only find discussions of how to indicate the distribution of the subject to conjuncts. The distribution of the subject is also indicated in the argument-value matrix on the right in the illustration above, namely, in that the first verb’s SUBJ(ect) is connected by a line to the second verb’s empty SUBJ slot.

The question I now have is whether connecting items this way is possible also for other features, like ASP(ect). From what little I know, I don’t see why it shouldn’t be plausible to analogize here, but I might as well be wrong. If you know, please let me know as well. What is slightly frustrating is that a lot of times, you can only easily find information on English.

Also, I’ve been working on writing this grammar for almost a whole year now. Wow.

  • Belyaev, Oleg, et al. “Number Mismatches in Coordination: An LFG Analysis.” Proceedings of the LFG ’15 Conference, Tokyo, Japan, 18–20 Jul. 2015. Ed. by Miriam Butt and Tracy Holloway King. Stanford, CA: CSLI Publications, 2015. 26–46. Web. 25 May 2017. ‹https://web.stanford.edu/group/cslipublications/cslipublications/LFG/20/papers/lfg15belyaevetal.pdf›.
  • Bresnan, Joan et al. Lexical-Functional Syntax. 2nd ed. Chichester: Wiley Blackwell, 2016. Print. Blackwell Textbooks in Linguistics 16.
  • Dalrymple, Mary. Lexical Functional Grammar. San Diego, CA: Academic Press, 2001. Print. Syntax and Semantics 34.
  • Kaplan, Ronald M., and John T. Maxwell, III. Constituent Coordination in Lexical-Functional Grammar. Palo Alto, CA: Xerox PARC, 1988. Web. 18 Apr. 2017. ‹http://www2.parc.com/isl/groups/nltt/xle/coord.ps›.
  • Maxwell, John T., III, and Christopher D. Manning. “A Theory of Non-constituent Coordination Based on Finite-State Rules.” Proceedings of the LFG ’96 Conference, Rank Xerox, Grenoble, France, 26–28 Aug. 1996. Ed. by Miriam Butt and Tracy Holloway King. Stanford, CA: CSLI Publications, 1997. Web. 18 Apr. 2017. ‹https://web.stanford.edu/group/cslipublications/cslipublications/LFG/1/lfg96maxwellmanning.pdf›.
  • Peterson, Peter G. “Coordination: Consequences of a Lexial-Functional Account.” Natural Language & Linguistic Theory 22.3 (2004): 643–679. Web. 18 Apr. 2017. ‹http://www.jstor.org/stable/4048099›.

Some Further Thoughts on Agreement in Ayeri

Ayeri makes use of grammatical gender to at least some degree. As a category, gender is by far not as pervasive as in Romance languages, though, since in Ayeri it only matters in picking the correct case suffix for agents and patients (animate vs. inanimate) as well as in 3rd person pronouns and verb agreement (masculine/feminine/neuter animate vs. inanimate). The fact that gender plays a role in verb agreement opens up possibilities for adding some quirks. As pointed out in a previous blog article,

Ayeri usually exhibits verbs as agreeing with agents and occasionally patients, not topics as such. This may be a little counterintuitive since the relation between topics and subjects is close, but is possibly due to the fact that the unmarked word order is VAP. This means that agent NPs usually follow the verb, and it strikes me as not too unnatural to have an agreement relation between the verb and the closest NP also when non-conjoined NPs are involved (Corbett [2006:] 180). This conveniently explains why verbs can agree with patients as well if the agent NP is absent.

This brings up the idea that Ayeri should likely exhibit agreement with the closest conjunct when a word is forced to agree with coordinated NPs of different genders, instead of finding some way to resolve conflicting gender features. Yet, however, I have often been following the rule that a masculine entity and a feminine entity, regardless of number, resolve to masculine as the default gender, though sometimes I have also used the neuter as a third category to escape to (Wechsler 2009 actually quotes Icelandic as following this route):

M+F=M(N attested)
F+M=M(N attested)

I was originally wondering if I should get rid of this system and instead use nearest-conjunct agreement throughout, but—why not have both? In this little blog article I basically want to sketch out an idea I had in the hope that it is well within the confines of what natural languages do. Note that the tables given in the following refer strictly to verb agreement, where the verb precedes any NPs it agrees with. Thus, for the outcome of nearest-conjunct agreement, it is assumed that the agreement target precedes the agreement controller, so that the verb should in most cases agree with the first conjunct.

Now, if NPs referring to people or other entities to which masculine and feminine apply as grammatical categories have their diverging gender features resolved to masculine as the default, what about combining either masculine or feminine with neuter? This is shown in the following table:

F+N=M(F possible)

From this table we can gather that in general, there is nearest-conjunct agreement for combinations of masculine/feminine and neuter, though feminine and neuter equally resolve to masculine as the default if the feminine conjunct is closer to the agreement target. Note that for agreement with simplex NPs, there is no default gender, so masculine gender will trigger masculine agreement, feminine gender will trigger feminine agreement, etc. As indicated in the table above, true nearest-conjunct agreement with the feminine conjunct is possible as well, however, basically ignoring any further complications. It takes not a lot of imagination to assume that the regularization towards nearest-conjunct agreement would be a point of divergence between the formal and the colloquial language and also very likely a fertile ground for prescriptivist bickering.

So far, we have only looked at the combinations within the animate tier. The following table lists the possible permutations for combinations of animate and inanimate NPs:

F+INAN=M(F possible)

For consistency, the same rules as above operate here: masculine and feminine mixed with inanimate show nearest-conjunct agreement. If, however, a feminine conjunct comes first, agreement will default to masculine, though again, nearest-conjunct agreement to feminine is possible.

Wechsler (2009: 571–73) furthermore discusses Corbett (1991)’s observation that there may be differences in how languages go about gender resolution with regards to semantic and syntactic resolution—i.e. resolution of conflicts in gender between the semantics and the form of a word—and finds that animacy plays a role in that. This is relevant in cases where grammatical and semantic gender diverge, as in hybrid nouns like German Mädchen ‘girl’. Mädchen semantically refers to a female person but by its form is of neuter gender, since the diminutive suffix -chen always derives neuter nouns. The question now is, which gender do pronouns and agreement referring to the girl have, neuter or feminine? In fact, variation can be observed in these cases.

While Ayeri assigns masculine and feminine semantically (with neuter for the remainder of animate entities which are neither male nor female), there are occasional idiosyncrasies with nouns very obviously referring to non-living things being assigned animate neuter gender, such as nanga ‘house’. It might be interesting to develop some further ideas for likely outcomes in that regard even though canonically, nearest-conjunct agreement should operate in those cases and conlangs, by their nature, probably produce a lot less variation than actual natural languages do.

  • Corbett, Greville G. Agreement. Cambridge: Cambridge UP, 2006. Print. Cambridge Textbooks in Linguistics.
  • ———. Gender. Cambridge: Cambridge UP, 1991. Print. Cambridge Textbooks in Linguistics.
  • Wechsler, Steven. “‘Elsewhere’ in Gender Resolution.” The Nature of the Word: Studies in Honor of Paul Kiparsky. Ed. by Kristin Hanson and Sharon Inkelas. Cambridge, MA: MIT Press, 2009. Print. 567–586. Current Studies in Linguistics.
  • I can’t seem to keep my terminology straight: I corrected *nearest-conjunct resolution to nearest-conjunct agreement in a few places.

Markov-Chain Generator for Ayeri Words

Since I’m sometimes a little lazy to come up with new words, I wrote myself a little Python script which pulls a certain subset of words from the dictionary database I’m using and applies a Markov chain algorithm to it in order to generate new similar words. The script is sophisticated enough to filter out duplicates and some other undesirable outcomes. You can adapt the code shared below to your needs if you wish to.