Comments on Test #2
Comments on Test #2Class Average: 41.5; Range: 15.5 to 54.
Hurrah!
Overall, almost everybody has something to be pleased about with this test. Average scores were 10% improved over the last test, and most individuals scored higher. Most of the questions were answered well by most people, with the exception of just two questions -- the questions on ambiguity and on syntactic structures in an unfamiliar language. Warning: the final exam will most probably include questions of this kind, so you should give extra attention to these when preparing for the final exam.
[5 points: 0.5 per answer]
This question presented few problems. The the main problem that arose was that many people gave justifying words containing an affix that sounds/is spelled the same as the affix in the example, but is not a case of the same morpheme, i.e. combines with quite different categories, and has quite different meaning. E.g. 'in-' in 'insincere' combines with adjectives, and has negative meaning; the 'in-' in 'inflate' or 'intent' does not have either of those properties. Similarly, '-th' in 'growth' combines with a verb to form a noun; the '-th' in 'fourth', 'tenth' does not have either of these properties, therefore it must be a different morpheme.
[4 points: 1 per answer]
(a) False, (b) True, (c) False, (d) True. Questions (a), (c), and (d) were covered in class lectures, question (b) was covered in the readings in Pinker's book required for Homework #9.
[4 points: 0.5 off for each error]
Mostly done fairly well, but with a number of fairly consistent problems.
[4 points: 1 for (i), 2 for (ii), 1 for (iii)]
(i) when a word formation rule was given, it was generally given correctly: combines with adjectives to create adverbs. If you did not correctly identify the categories involved, then this is something that you should make it a priority to learn ... as this is something very basic. Many people described spelling conventions associated with the suffix, which are irrelevant.
(ii) Mostly done well, with examples. More than 2 examples of possible combinations are required to conclude productivity. Many people found impossible examples ('funly', 'bluely', 'fastly'). An alternative test of productivity (used by few people) involved testing combinations with novel words.
(iii) Very few problems.
[8 points: 5 points for (a), 1 for (b), 2 for (c)]
Almost universal high scorer. Very few errors in (a) or (c). (b) required mention of both a possessive prefix and a number suffix; just giving a list of the prefixes or suffixes is not enough - there is a generalization that all the prefixes, for example, have a similar function.
[5 points: 0.5 off per error]
Most trees were quite good. Main problems were (i) failing to embed 'that the first cat ...' inside the main clause, (ii) misplacing or mislabeling the Auxiliary in embedded clause - should be a daughter of the embedded S.
[5 points: 2 points for meanings, 2 points for structures, 1 point for relating the ambiguity to the NP structure]
Many problems here, most of which should have been easily avoidable.
(i) Meanings. Many of the meanings given simply do not correspond to interpretations of the sentence 'the man with the bicycle from Italy arrived'. This does not mean the same thing as 'the man with the bicycle arrived from Italy', as any speaker of English can quickly verify. Do not give paraphrases which mean something that the example sentence does not mean.
(ii) Tree structures. The point of structural ambiguity is that one sequence of words can have multiple meanings. Therefore, giving tree structures for two different sequences of words is irrelevant. This includes giving a tree structure for your paraphrase of one of the meanings.
(iii) Tree structures. The ambiguity involves whether the second PP 'from Italy' is a modifier of the first NP 'the man' or the second NP 'the bicycle'. Very many people attached the PP-modifiers to the VP rather than to the NP.
Given that structural ambiguity is discussed in the textbook, was covered in one homework and was discussed in multiple classes and discussion sections, this should not have been too challenging.
[5 points: 3 points for rules, 2 points for tree for (c)]
Again many problems here.
It is possible to create phrase structure rules for the Kannada sentences using exactly the phrase structure rules that we have used in English, but changing the order of some of the elements (e.g. PP --> NP P, instead of the English PP --> P NP). However, there are elements that often occur in English which do not occur in Kannada, e.g. determiners. Therefore, there is no need for rules like NP --> Det N.
In diagramming (c), very many people put the NP tigane/bedbug as the sister of the preposition inda/with. This should have created warning signals: the translation of the sentence indicates that something was done with a finger, not with a bedbug!
Similarly, a number of people appropriately treated the P 'inda' as attached inside the verb phrase, but incorrectly attached the noun meaning "finger" as a part of the subject noun phrase. By paying attention to what the meaning of the sentence tells us about which words should be grouped together, it should have been possible to avoid this problem.
[3 points: 1 for critical test, 1 for argument/modifier diagnosis, 1 for tree structure]
This question was answered very well overall. Many people considered "on the table" to be a modifier, and less than I expected considered it to be an argument. Both of these answers were given full credit if the logic of the answer was correct and the associated tree was correct.
We had expected that everybody would consider "on the table" in "he dropped the book on the table" to be an argument, but the difference in opinions may have been due to an ambiguity in what "on the table" means.
If you find the following acceptable:
Are you interpreting "on the table" and "on the couch" as the location of John and Sue, or as the place where the book was dropped to? I would expect that the only available interpretation would be the one in which the PP describes the location of the subjct NP.
The following sentence disambiguates, and only allows the target-of-dropping interpretation.
If you found that "on the table" was a modifier, do you find that "into the chute" behaves as an argument in the following?
[6 points: 3 points for frames, 3 points for appropriate justifications]
[6 points: 2 for environments, 1 for minimal pairs, 1 for phoneme/allophone, 2 for rule/generalization]
Overall well done, in the listing of environments and the noticing of the appropriate generalization. However, a number of people slipped up unnecessarily, by making exactly the right generalization about the complementarity of b/p in this language, but nevertheless claiming that this is evidence that these two sounds are separate phonemes in this language.