Íslenskt mál og almenn málfræði - 01.01.2008, Side 70

68
Ida Larsson
In this respect, the construction with vera búinn að behaves like the
resultative:
(30) a. *Ef hann væri kominn í gær, þá...
if he were come yesterday then
b. Ef hann væri kominn núna, þá...
if he were come now then
‘If he were here now, then...’
Similar pattems can be observed in the other Germanic languages.
The auxiliary be appears to be possible in past tense counterfactuals
only in languages that form perfects with be. Hence, in Danish, the
temporal auxiliary vœre ‘be’ is possible in past counterfactuals (cf.
McFadden and Alexiadou 2007):
(31) Jeg havde hábet pá at han var kommet til modet i gár.
I had hoped on that he was come to the meeting yesterday
‘I had hoped that he had come to the meeting yesterday.’
The passive auxiliary vœre, on the other hand, cannot have a past
tense reading in a counterfactual, unless it is embedded under a per-
fect:13
(32) a. *Hvis trojen var vasket til festen igár,
if the shirt was washed to the party yesterday
ville jeg have bmgt den da.
would I have used it then
Intended reading: ‘If the shirt had been washed for the party
yesterday, I would have wom it then.’
b. Jeg havde hábet pá at trajen havde været vasket i gár.
I had hoped on that the shirt had been washed yesterday
‘I had hoped that the shirt had been washed yesterday.’
13 In languages such as Old Swedish and Old and Middle English, the auxiliary
be (vara) + active participle was impossible in past counterfactual contexts (see e.g.
Johannisson 1945, Rydén and Brorström 1987, Larsson 2007); this suggests that be
was not a temporal auxiliary (cf. McFadden and Alexiadou 2006a, 2006b, 2007).
Similarly, when the perfect with be fírst developed in German and Dutch, there were
restrictions on be in counterfactuals (see Shannon 1995 and Ledgeway 2003).