2. Previous Research
2.1. Typological Studies
2.2.7. Noonan (1994)
55
‘Soso sent his (own) sister a present.’
[example and translation from Orr (1989: 13); detransliteration mine]
There are some problems concerning Russian (of which discussion is omitted in this dissertation), but he concludes that the construction with the preposition у ‘at’ in northern Russian, like in (144), and the preposition ag ‘at’ in Munster (southern) and Connacht (western) Irish, like in (145), can be classified as the ergative construction.
144) У него своя изба поставлена.
u n-ego svoj-a izb-a po-stavl-ena
at E-3SG.M.GEN REFL.POSS-F.NOM cottage.F-SG.NOM PFV-place-PTCP.PASS.PST
‘He has placed his own cottage.’
[example from Orr (1989: 10); detransliteration mine]
145) Tá sé déanta agam.
tá-Ø sé déan-ta ag-am be-PRS 3SG.M.CNJTV do-VA at-1SG
‘I have done it.’
[example and translation from Orr (1989: 6)]
56 147) Passive (progressive):
Bhí Seán á bhualadh (ag Liam).
L bí-Ø Seán-Ø do+aL bual-adh ag Liam-Ø
PM be-PST Seán.M-SG.NOM to+3SG.M.POSS hit-VN at Liam.M-SG.NOM
‘Seán was being hit (by Liam).’
[example from Noonan (1994: 280)]
148) Active (simple tense):
Bhuail Liam Seán.
L buail-Ø Liam-Ø Seán-Ø
PM hit-PST Liam.M-SG.NOM Seán.M-SG.NOM
‘Liam hit Seán.’
[example from Noonan (1994: 280)]
149) Passive (impersonal passive):
Bualadh Seán (le Liam).
bual-adh Seán-Ø le Liam-Ø hit-PST.IMPERS Seán.M-SG.NOM with Liam.M-SG.NOM
‘Seán was hit (by Liam).’
[example from Noonan (1994: 280)]
Noonan (1994) considers the topicality of each argument in some constructions in Irish (including the
‘be done’ construction) based on 400 clauses of spoken Irish and 400 clauses of written Irish. Note the two specific terms, Referential Distance (RD) and Persistence (Per). The former is a count of the number of clauses since the last reference (max: 20)27 and the latter is the number of times it appears in the following ten clauses. According to Noonan, ‘A low value for referential distance correlates with high topicality; a high value for persistence correlates with high topicality’ [Noonan (1994: 308)].
First, Table 15 shows the topicality of agents and patients in verbal clauses (clauses with finite verbs, including the impersonal passive).
27 Any new reference (e.g. an indefinite) is assigned a referential distance of twenty [Noonan (1994: 308)].
57 Table 15: Topicality of Agents and Patients in Verbal Clauses
spoken data written data
RD Per n RD Per n
impersonal
patient 6.61 2.52 32 8.76 2.43 52
intransitive
subject 6.99 2.31 39 9.00 2.22 46
transitive
subject 6.33 3.02
85 5.67 2.98 direct object 12.97 1.40 14.76 0.35 54
[based on Noonan (1994: 291)]
Concerning the referential distance, that of the patient of an impersonal passive (6.61 and 8.76) and that of the intransitive subject (6.99 and 9.00) are very similar, but that of the transitive direct object (12.97 and 14.76) is quite different. Similarly, concerning persistence, that of the impersonal patient (2.52 and 2.43) and that of the intransitive subject (2.31 and 2.22) are very similar, but that of the transitive direct object (1.40 and 0.35) is quite different. This suggests that patients of an impersonal passive, the intransitive subject and the transitive subject have high topicality.
Table 16 shows the topicality of agents and patients in nominal clauses (clauses composed of verbal nouns, including the passive in his term).
Table 16: Topicality of Agents and Patients in Nominal Clauses spoken data written data
RD Per n RD Per n
passive
agent 3.27 2.95
22 4.83 5.83 12
patient 16.55 0.45 19.33 0.67
passive (agentless)
patient 20.00 0.50 2 20.00 0.00 6
active-transitive
agent 13.93 1.43
14 2.00 3.87
patient 7.23 2.54 12.20 1.27 30
active-intransitive
agent 7.73 3.64 11 10.50 5.00 4
[based on Noonan (1994: 293)]
Concerning the passive, the referential distance of a patient (16.55 and 19.33) is higher than that of an agent (3.27 and 4.83), and the persistence of a patient (0.45 and 0.67) is lower than that of an agent (2.95 and 5.83). Then, comparing passive patients and active-transitive patients, the referential distance of the former (16.55 and 19.33) is higher than that of the latter (7.23 and 12.20), and the persistence of the
58
former (0.45 and 0.67) is lower than that of the latter (2.54 and 1.27). This result suggests that passive patients have low toplicality.
Based on this statistical data, he concludes that the passive conforms to the structural characterisation of passive but fails to conform to the functional characterization, i.e. syntactic features such as promotion and demotion of arguments can be observed, but their information structures are not those of the passive.
In contrast, he says that the impersonal passive fails in the structural characterisation of the passive but conforms to the functional characterization, i.e. syntactic features such as promotion and demotion of arguments cannot be observed28, but its information structure is that of the passive.
Then, in Chapter Four, ‘The Information Structure of Irish Sentences’, he compares the information structure of the two passives above and that of other sentences. One of them is the ‘be done’ construction (called participial clauses, as opposed to nominal clauses with verbal nouns like (146) and (147)).
He says that participial clauses in Irish display a straightforward ergative syntax, and their sense is resultative/perfect. In other words, in simple tense constructions, like those in (150), (152) and (154), an agent appears as a grammatical subject, and a patient of a transitive clause appears as a grammatical object.
In contrast, in ‘be done’ constructions, like those in (151), (153) and (155), an agent appears as a grammatical subject in intransitive clauses, but a patient appears as a grammatical subject in transitive clauses, and there an agent is expressed by the preposition ag ‘at’ (if any).
150) simple tense: intransitive
D’imigh na mic léinn abhaile ar saoire na Cásca.
do im-igh na mac-i léann-i abhaile ar saoire-Ø na
PM go.out-PST DEF.PL.NOM son.M-PL.NOM learning.M-SG.GEN home on feast.F-SG.NOM DEF.F.GEN
Cásc-a Easter.F-SG.GEN
‘The students went home for Easter.’
[developed from an example by Noonan (1994: 297)]
151) ‘be done’: intransitive
Tá na mic léinn imithe abhaile ar saoire na Cásca.
tá-Ø na mac-i léann-i im-ithe abhaile ar saoire-Ø be-PRS DEF.PL.NOM son.M-PL.NOM learning.M-SG.GEN go.out-VA home on feast.F-SG.NOM
na Cásc-a
DEF.F.GEN Easter.F-SG.GEN
‘The students had gone home for Easter.’
[example and translation from Noonan (1994: 297)]
28 In the impersonal passive, the patient is expressed as a grammatical object (see §1.5.3.).
59 152) simple tense: transitive
Níor fhág siad pána gloine sa teach.
ní+roL fág-Ø siad pána-Ø gloine-Ø i+an teach-Ø
NEG+PM leave-PST 3PL.CNJTV pane.M-SG.NOM glass.F-SG.GEN in+DEF.M.NOM house.M-SG.NOM
‘They did not leave a pane of glass in the house.’
[developed from an example by Noonan (1994: 297)]
153) ‘be done’: transitive (agentless) Ní raibh pána gloine fágtha sa teach.
ní raibh-Ø pána-Ø gloine-Ø fág-tha i+an teach-Ø
NEG be-PST pane.M-SG.NOM glass.F-SG.GEN leave-VA in+DEF.M.NOM house.M-SG.NOM
‘Not a pane of glass was left in the house.’
[example and translation from Noonan (1994: 297)]
154) simple tense: transitive
Chonaic an garraíodóir an gadhar.
L conaic-Ø an garraíodóir-Ø an gadhar-Ø
PM see-PST DEF.M.NOM gardener.M-SG.NOM DEF.M.NOM dog.M-SG.NOM
‘The gardener saw the dog.’
[developed from an example by Noonan (1994: 297)]
155) ‘be done’: transitive
Bhí an gadhar feicthe ag an ngarraíodóir.
L bí-Ø an gadhar-Ø feic-the ag anE garraíodóir
PM be-PST DEF.M.NOM dog.M-SG.NOM see-VA at DEF.M.NOM gardener.M-SG.NOM
‘The dog was seen by the gardener.’
[example and translation from Noonan (1994: 297)]
The statistical data regarding the topicality of agents and patients in participial clauses (= the ‘be done’ construction in my term) is given below.
Table 17: Topicality of Agents and Patients in Participial Clauses spoken data written data
RD Per n RD Per n
transitive
agent 3.33 2.89
9 3.27 4.91
patient 14.00 0.44 14.62 0.91 11
transitive (agentless)
patient 12.00 0.88 7 13.83 0.40 6
intransitive
agent 0 11.25 0.75 4
[based on Noonan (1994: 298)]
As shown in the table, transitive agents (expressed by the preposition ag ‘at’) in the ‘be done’
60
construction have high topicality as the referential distance is low (3.33 and 3.27) and the persistence is high (2.89 and 4.91). In contrast, transitive patients and intransitive agents (both are grammatical subjects) in the ‘be done’ construction have low topicality as the refential distance is high (14.00 and 14.62 in the former; 11.25 in the latter for the written data) and the persistence is low (0.44 and 0.91; 0.75).
Although he does not specifically mention it, it can be assumed that this is the reason why he calls the construction concerned ‘ergative’, not ‘passive’.