From: Oystein Nilsen <oystein.nilsenlet.uu.nl>
Subject: English Judgments on the Meaning of 'other'
Query for this summary posted in LINGUIST Issue:
This is a summary of the query posted here:
We received 74 completed forms, 9 of which we excluded from
consideration (we excluded forms where sentences 22 and 26 were judged
'good') thus leaving us with 65 completed forms. This happened within
about 24 hours after the form appeared on the linguist list. We are deeply
grateful to all those who responded, and also quite surprised at how
effective this way of gathering judgments turned out to be.
There were three conditions, each with two or more variables. First, we
tested bound readings of 'other' with a variety of different quantifiers as
binders, i.e. quantifiers headed by 'no,' 'all,' 'every,' 'the,' 'and,' 'two,' and
Secondly, we tested whether bound readings of 'other' are sensitive to
syntactic locality. We tested clausmate binding, cases where the binder and
'other' were separated by a finite clause boundary, a 'small clause' (ECM)
boundary, or by a factive island.
Finally, we tested whether there is an effect on the availability of bound
readings of the presence/absence of a noun following 'other' (i.e. 'the
others' vs. 'the other students').
A table summarizing all our findings can be downloaded here:
and the original form (no longer active) can be found here:
Some of the effects turned out not to be significant. For example, sentences
in the clause mate binding configuration, with 'the' as the head of the
binder were accepted by 22% when 'other' was not followed by a noun, and
34% of the time when it was. But this difference was not significant (s=.1).
We have not completed significance testing of all the pairs we'd like to, but
when we have, we'll post it where the table can be found now.
If somebody is interested in the raw material, we can email it to them if they
We received several comments, not all of which we can respond to here. One
recurrent remark was that sentences with 'most' are infelicitous in a context
that only has three individuals. We are planning a follow-up study where we
repair this defect. Another frequent remark was that the word ''felicitous'' is
somewhat unfortunate. In our follow-up study, we will use locutions like
'make sense' instead. A third point with which we also agree, is that the
GOOD/BAD dichotomy is too coarse for some of these judgments. Hence we
are planning to have more gradients in our follow-up study. This will also
give us more statistical power, and potentially lead to more significant
Jakub Dotlacil and Øystein Nilsen
Respond to list|Read more issues|LINGUIST home page|Top of issue