From mcginnis at ucalgary.ca Mon Oct 7 17:05:51 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Mon, 7 Oct 2002 11:05:51 -0600 Subject: Alexandra Galani: copies of papers Message-ID: Dear all, I found the papers I needed. Thank you for all your help! Alexandra From mcginnis at ucalgary.ca Fri Oct 25 20:47:03 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Fri, 25 Oct 2002 14:47:03 -0600 Subject: Mark Volpe: ROOT in DM Message-ID: Hi Listers, The term ROOT as I've understood is "what remains when all morphological information is wrung out of a form" (Aronoff, 1994: 40). Is this the definition which is assumed in Marantz (1997)? Based on English examples, where ROOTs and Stems are isomorphic, it's not clear what a DM ROOT must be, Moreover, to my knowledge, I've never seen it defined. This may be a good thing. Based on my research, the strict definition of ROOT is inadequate to the task. Let me try to provide in example. In Heidi Harley's wonderful dissertation (1995a) and in an unpublished paper (1995b) she develops an elegant system which seems adequate to handling the Lexical Syntax of the Causative Alternation in Japanese. The unaccusative is base-generated under the head of VP with the abstract morpheme BECOME, the lexical causative is base generated under the head of vP (Heidi's EventP) under the abstract morpheme CAUSE. In these cases you have a bi-unique relation between BECOME and unaccusative, CAUSE and lexical causative. However, important facts about lexical causatives in Japanese have not been noted in the extant literature, e.g. Miyagawa (1989) and Kuroda (1993). In fact, lexical causatives are paired with unergatives, so that in Japanese you can 'run your friend to the liquor store', 'play your children in the park', 'sleep your baby in the afternoon', etc. Moreover, lexical causatives are paired with transitives so that you get 'look at/ show', 'eat/ feed', 'put on'/dress', etc. Now the bi-unique relationship between the alternations and Lexical Syntax has dissolved. Unergative/ Transitive paired with lexical causative are both neccessarily base-generated under head of vP with the abstract morpheme CAUSE. A ROOT under the strict definition cannot disambiguate one from the other. Has anyone been playing with the "definition" of ROOT or found need to in their research. My intuition is that to the extent that it remains categorically unspecified, it should be a DM ROOT. I look forward to any comments. Mark __________________________________________________ Do you Yahoo!? Y! Web Hosting - Let the expert host your web site http://webhosting.yahoo.com/ From mcginnis at ucalgary.ca Mon Oct 28 16:14:55 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Mon, 28 Oct 2002 09:14:55 -0700 Subject: Dan Everett: ROOT in DM (reply to Mark Volpe) Message-ID: Folks, Mark Volpe's posting was interesting. I can see his point, but there could be evidence for ROOT which he, if I understand correctly, overlooks. Consider eat/feed, die/kill, and so on. The elimination of ROOT for these pairs would seem to be based on the assumption that all the differences between them are captured by the syntactic structures that derive them. But if there were other meaning/usage distinctions that did not involve, say, CAUSE, but instead were cultural, idiomatic, etc, then these would arguably emerge from different ROOTs. Related to this is the question as to whether languages with overt causative morphology distinguish three-way contrasts such as CAUSE+die vs. die vs. (a separate word) 'kill', and so forth. This would mean that individual members of derivational pairs might possess nonderivational and non-eliminable residues. ROOT is ideal for capturing these distinctions. Dan ************************************************************** Daniel L. Everett Professor of Phonetics and Phonology University of Manchester Oxford Road Manchester, UK M13 9PL dan.everett at man.ac.uk Office Phone: 44-161-275-3158 Department Phone & Fax: 44-161-275-3187 From mcginnis at ucalgary.ca Mon Oct 28 18:15:35 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Mon, 28 Oct 2002 11:15:35 -0700 Subject: Mark Volpe: ROOT in DM (reply to Dan Everett) Message-ID: Hi Listers, Thanks Dan! I think Haspelmath has observed that languages, in general, avoid deriving 'die' from the same ROOT as 'kill'. This is certainly a cultural aversion, to some extent. The fact that the human race seems to recognize 'death' as an internally-caused event, distinct from 'murder' is relevant, as well. It's quite true, as you observe, at least for Japanese, that while 'die' and 'cause to die' are from the same ROOT, 'kill' is not. My point was that where 2 vocabulary items are both agentive, e.g., 'miru' ('look at'), and 'miseru' ('show'), a strict ROOT, 'mi-' will not be able to distinguish which vocabulary item is being selected. I must tell you that I've had the very good fortune of discussing the issues with Heidi Harley, and I believe we have come to a consensus on a DM-friendly resolution. My concern is that, if it is only Japanese which needs such a solution, it would seem less motivated. BTW, Dan, I'm a big fan of your piece on 'Wari' in Handbook of Morphology'. That is you, isn't it? Thanks,---Mark __________________________________________________ Do you Yahoo!? Y! Web Hosting - Let the expert host your web site http://webhosting.yahoo.com/ From mcginnis at ucalgary.ca Tue Oct 29 16:15:24 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Tue, 29 Oct 2002 09:15:24 -0700 Subject: Dan Everett: ROOT in DM (reply to Mark Volpe) Message-ID: Dear Mark, I suspect that the problem is richer than we realize, but I take your point. (Of course, I would take anyone's point when they praise an article of mine. If you liked the Wari' chapter in the Handbook of Morphology, you might want to check out my piece on the RRG website which, eventually, will make its way into a journal.) But it would be interesting to hear if others know of 3-way contrasts between sets like 'kill', 'die', and 'CAUSE+die'. Dan ******************** Dan Everett Professor of Phonetics and Phonology Department of Linguistics University of Manchester Oxford Road Manchester, UK M13 9PL Phone: 44-161-275-3158 Department Fax: 44-161-275-3187 From mcginnis at ucalgary.ca Tue Oct 29 17:49:39 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Tue, 29 Oct 2002 10:49:39 -0700 Subject: Andrew Nevins: Shmurvey Message-ID: Dear DM Enthusiasts, We would like to invite you to take a few moments to complete a survey on your intuitions on English "shm-" reduplication (e.g. "lexicalism, shmexicalism"). Should you be kind enough to help, or have any thoughts to share, the URL is: http://www.ai.mit.edu/projects/dm/shm Thanks! Andrew Nevins (MIT): anevins at mit.edu Bert Vaux (Harvard): vaux at fas.harvard.edu From mcginnis at ucalgary.ca Thu Oct 31 16:51:16 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Thu, 31 Oct 2002 09:51:16 -0700 Subject: Correction: EPP/phase workshop Message-ID: Hi all, I goofed! The address to send abstracts to for January's EPP/Phase workshop at MIT is marthajo at alum.mit.edu, NOT mcginnis at alum.mit.edu, as announced on LINGUIST. (Mark McGinnis is not going to be pleased with me!) If you are submitting an abstract or know anyone else who is, please take note of the corrected address. I've sent out a correction to LINGUIST too. The deadline is, of course, today. Happy Hallowe'en, Martha -- mcginnis at ucalgary.ca From mcginnis at ucalgary.ca Mon Oct 7 17:05:51 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Mon, 7 Oct 2002 11:05:51 -0600 Subject: Alexandra Galani: copies of papers Message-ID: Dear all, I found the papers I needed. Thank you for all your help! Alexandra From mcginnis at ucalgary.ca Fri Oct 25 20:47:03 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Fri, 25 Oct 2002 14:47:03 -0600 Subject: Mark Volpe: ROOT in DM Message-ID: Hi Listers, The term ROOT as I've understood is "what remains when all morphological information is wrung out of a form" (Aronoff, 1994: 40). Is this the definition which is assumed in Marantz (1997)? Based on English examples, where ROOTs and Stems are isomorphic, it's not clear what a DM ROOT must be, Moreover, to my knowledge, I've never seen it defined. This may be a good thing. Based on my research, the strict definition of ROOT is inadequate to the task. Let me try to provide in example. In Heidi Harley's wonderful dissertation (1995a) and in an unpublished paper (1995b) she develops an elegant system which seems adequate to handling the Lexical Syntax of the Causative Alternation in Japanese. The unaccusative is base-generated under the head of VP with the abstract morpheme BECOME, the lexical causative is base generated under the head of vP (Heidi's EventP) under the abstract morpheme CAUSE. In these cases you have a bi-unique relation between BECOME and unaccusative, CAUSE and lexical causative. However, important facts about lexical causatives in Japanese have not been noted in the extant literature, e.g. Miyagawa (1989) and Kuroda (1993). In fact, lexical causatives are paired with unergatives, so that in Japanese you can 'run your friend to the liquor store', 'play your children in the park', 'sleep your baby in the afternoon', etc. Moreover, lexical causatives are paired with transitives so that you get 'look at/ show', 'eat/ feed', 'put on'/dress', etc. Now the bi-unique relationship between the alternations and Lexical Syntax has dissolved. Unergative/ Transitive paired with lexical causative are both neccessarily base-generated under head of vP with the abstract morpheme CAUSE. A ROOT under the strict definition cannot disambiguate one from the other. Has anyone been playing with the "definition" of ROOT or found need to in their research. My intuition is that to the extent that it remains categorically unspecified, it should be a DM ROOT. I look forward to any comments. Mark __________________________________________________ Do you Yahoo!? Y! Web Hosting - Let the expert host your web site http://webhosting.yahoo.com/ From mcginnis at ucalgary.ca Mon Oct 28 16:14:55 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Mon, 28 Oct 2002 09:14:55 -0700 Subject: Dan Everett: ROOT in DM (reply to Mark Volpe) Message-ID: Folks, Mark Volpe's posting was interesting. I can see his point, but there could be evidence for ROOT which he, if I understand correctly, overlooks. Consider eat/feed, die/kill, and so on. The elimination of ROOT for these pairs would seem to be based on the assumption that all the differences between them are captured by the syntactic structures that derive them. But if there were other meaning/usage distinctions that did not involve, say, CAUSE, but instead were cultural, idiomatic, etc, then these would arguably emerge from different ROOTs. Related to this is the question as to whether languages with overt causative morphology distinguish three-way contrasts such as CAUSE+die vs. die vs. (a separate word) 'kill', and so forth. This would mean that individual members of derivational pairs might possess nonderivational and non-eliminable residues. ROOT is ideal for capturing these distinctions. Dan ************************************************************** Daniel L. Everett Professor of Phonetics and Phonology University of Manchester Oxford Road Manchester, UK M13 9PL dan.everett at man.ac.uk Office Phone: 44-161-275-3158 Department Phone & Fax: 44-161-275-3187 From mcginnis at ucalgary.ca Mon Oct 28 18:15:35 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Mon, 28 Oct 2002 11:15:35 -0700 Subject: Mark Volpe: ROOT in DM (reply to Dan Everett) Message-ID: Hi Listers, Thanks Dan! I think Haspelmath has observed that languages, in general, avoid deriving 'die' from the same ROOT as 'kill'. This is certainly a cultural aversion, to some extent. The fact that the human race seems to recognize 'death' as an internally-caused event, distinct from 'murder' is relevant, as well. It's quite true, as you observe, at least for Japanese, that while 'die' and 'cause to die' are from the same ROOT, 'kill' is not. My point was that where 2 vocabulary items are both agentive, e.g., 'miru' ('look at'), and 'miseru' ('show'), a strict ROOT, 'mi-' will not be able to distinguish which vocabulary item is being selected. I must tell you that I've had the very good fortune of discussing the issues with Heidi Harley, and I believe we have come to a consensus on a DM-friendly resolution. My concern is that, if it is only Japanese which needs such a solution, it would seem less motivated. BTW, Dan, I'm a big fan of your piece on 'Wari' in Handbook of Morphology'. That is you, isn't it? Thanks,---Mark __________________________________________________ Do you Yahoo!? Y! Web Hosting - Let the expert host your web site http://webhosting.yahoo.com/ From mcginnis at ucalgary.ca Tue Oct 29 16:15:24 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Tue, 29 Oct 2002 09:15:24 -0700 Subject: Dan Everett: ROOT in DM (reply to Mark Volpe) Message-ID: Dear Mark, I suspect that the problem is richer than we realize, but I take your point. (Of course, I would take anyone's point when they praise an article of mine. If you liked the Wari' chapter in the Handbook of Morphology, you might want to check out my piece on the RRG website which, eventually, will make its way into a journal.) But it would be interesting to hear if others know of 3-way contrasts between sets like 'kill', 'die', and 'CAUSE+die'. Dan ******************** Dan Everett Professor of Phonetics and Phonology Department of Linguistics University of Manchester Oxford Road Manchester, UK M13 9PL Phone: 44-161-275-3158 Department Fax: 44-161-275-3187 From mcginnis at ucalgary.ca Tue Oct 29 17:49:39 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Tue, 29 Oct 2002 10:49:39 -0700 Subject: Andrew Nevins: Shmurvey Message-ID: Dear DM Enthusiasts, We would like to invite you to take a few moments to complete a survey on your intuitions on English "shm-" reduplication (e.g. "lexicalism, shmexicalism"). Should you be kind enough to help, or have any thoughts to share, the URL is: http://www.ai.mit.edu/projects/dm/shm Thanks! Andrew Nevins (MIT): anevins at mit.edu Bert Vaux (Harvard): vaux at fas.harvard.edu From mcginnis at ucalgary.ca Thu Oct 31 16:51:16 2002 From: mcginnis at ucalgary.ca (Martha McGinnis) Date: Thu, 31 Oct 2002 09:51:16 -0700 Subject: Correction: EPP/phase workshop Message-ID: Hi all, I goofed! The address to send abstracts to for January's EPP/Phase workshop at MIT is marthajo at alum.mit.edu, NOT mcginnis at alum.mit.edu, as announced on LINGUIST. (Mark McGinnis is not going to be pleased with me!) If you are submitting an abstract or know anyone else who is, please take note of the corrected address. I've sent out a correction to LINGUIST too. The deadline is, of course, today. Happy Hallowe'en, Martha -- mcginnis at ucalgary.ca