Here's another take on relative pronouns. I happened to be checking Curme's Parts of Speech and Accidence this morning, on another matter entirely, when I came across the following in a section headed "Relative Pronouns with Antecedent": "These relative pronouns are who, which, that, as, but, but that, but what (colloquial), the indefinites whoever, whatever, and whichever, and other less common forms enumerated in [his Syntax, the other part of Curme and Kurath's A Grammar of the English Language (HFWS]." I suspect we could get into an interesting discussion of "as", "but", and "but that". Herb ________________________________ From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Bruce Despain Sent: Monday, March 14, 2005 10:36 AM To: [log in to unmask] Subject: Re: which and that Herb, Thank you for going to the trouble to quote H&P so thoroughly. It is a convincing and persuasive argument! My mental block was in accepting the connective as a relative subordinator without the added baggage involved in pronominalization. Thinking of pronouns as the result of a process had trapped my thinking into supposing many, many varieties of the process, beyond what would actually be necessary for a concise description. BTW: the second sentence is a bit awkward logically. The use of that is in fact wider than the use of relative pronouns, and, thank goodness, is not conditional on it being a pronoun at all! Bruce >>> [log in to unmask] 3/12/2005 11:17:33 PM >>> Here's the H&P argument that relative that is not a pronoun (1056-7). I'll paraphrase where I can. "Wide range of antecedent types and relativized elements If that were a pronoun, its ue would be much wider than that of the uncontroversial relative pronouns, or indeed of any pro-form at all in language. Compare:" (sentences quoted, number and comparison formats changed from small Roman) 1. They gave the prize to the girl [that spoke first]. =who 2. Have you seen the book [that she was reading]? =which 3. He was due to leave the day [that she arrived]. =when 4. He followed her to every town [that she went]. =where 5. That's not the reason [that she resigned]. =why 6. I was impressed by the way [that she controlled the crowd]. =*how 7. It wasn't to you [that I was referring]. no wh form 8. She seems to be the happiest [that she has ever been]. no wh form . . . "Instead of proposing a pro-form with such an exceptional range of use, we are saying that that-relatives do not contain any overt pro-form linked to the antecedent; they simply have an anaphoric gap, like bare relatives." "Lack of upward percolation There are no that-relatives matching wh-relatives with a complex relative phrase: the woman [whose turn it was] *the woman [that's turn it was] the knife [with which he cut it] *the knife [with that he cut it]" If that were a pronoun we'd have to specify that it can't take a possessive and that it can't occur as the object of a preposition unless that preposition is stranded. These strict limitations contrast strongly with the versatility that (1-8) indicate. But if "that" is a subordinator, then these limitations are predicted naturally by the analysis: subordinators don't behave in either way. "Finiteness" That-relatives are always finite, as are the declarative content clauses introduced by "that". Thus the only bare relative that can't be introduced by "that" is the infinitival relative: a knife to cut it with vs. *a knife that to cut it with. This is predicted naturally by the conjunction analysis but requires a special rule under the pronoun analysis. "Omissibility" "That" is omissible in both relative clauses and content clauses. The conditions for deletion differ between the two, but in both structures omissibility is governed by the need to mark the beginning of the clause to avoid parsing problems. "And in both cases, 'that' is more readily omitted in simple structures than in complex ones. There is no pro-form in English that is systematically omissible under remotely similar conditions." That's their argument. Their first, third, and fourth arguments go beyond those that I've raised. Herb -----Original Message----- From: Assembly for the Teaching of English Grammar on behalf of Bruce D. Despain Sent: Sat 3/12/2005 9:11 AM To: [log in to unmask] Subject: Re: which and that Herb, You say that February must be coreferential to its head. Doesn't that make it a pronoun? My impulse was to analyze it allowing that the head was an article-like properness in the specific February. What I overlooked was a second complication of personification in the matrix, which may have affected the choice of connective. Let's try it with a definite noun (distorted content, however). (1) The treasurer, that in other companies held the responsibility to make deals, in this company kept funds tight. (2) The treasurer, who in other companies held the responsibility to make deals, in this company kept funds tight. In (1) the predicate of the matrix seems to be about the position, in (2) the person. The fact that in the dependent clause the referent is generic and in the matrix specific, does not seem to affect the choice of connective after all. So when we say (3), it seems a bit odd. We can't personify well with "which." (3) ?February, which in other years held intimations of spring, this year prolonged the bitter weather. With the other example of the supplementary relative we agreed on the antecedent. I think it is significant that the reference seems to be to instincts already mentioned previously, that indeed there is a demonstrative or at least pronominal aspect to the connective being used. The supplementary nature would then have to do with the missing antecedent as in (4). (4) She had long been accustomed to the solitary nature of her son's instincts, (the ones) that I had tried--and failed--to stifle. Bruce ----- Original Message ----- From: "Stahlke, Herbert F.W." <[log in to unmask]> To: <[log in to unmask]> Sent: Friday, March 11, 2005 7:04 PM Subject: Re: which and that Bruce, I do differ with you on your analysis of the February sentence. I can't see how February can be a common noun in the RC since it must be coreferential to its head. The only difference I see in the two uses is that the head is specific while the coreferent, which is 0, is generic. But it's proper in both cases. In the second sentence, I agree that not all the son's instincts are in play, just his solitary ones, but here I don't read a specific/generic contrast. They're both specific. I'm puzzled also by the domain of the reference of "that" being of article scope rather than of NP scope. I wonder if the demonstrative "that" is interfering here. But we remain pretty much apart on the issue. And I think the problem is a default claim that relative that is pronominal. That claim has to be argued for, given the close similarity of relative that, I would say identity, to conjunctive that. Herb Herb, Thank you for your patience in helping me out. Your points are well taken. I'm still a little uncomfortable about melding (2) and (3). The nature of the "pronoun" of (3) seems to turn on the restrictive/non-restrictive opposition. The two examples of non-restrictive clauses are problematic for me and may deserve a more careful analysis. Let me work this out. February, that in other years held intimations of spring, this year prolonged the bitter weather. In this sentence the name of the month is proper, if it is referring to the one month of the year with that name. When the author brings into consideration the other years, the noun becomes common. In the dependent clause the noun is common. In the main clause it is proper. The proper noun being fully defined in the main clause cannot be identified better by adding an adjective clause. So the clause is indeed non-restrictive by the normal definition of modification. However, the dependent clause is complementing the definiteness of the proper noun of the main clause. This is the same function that the content clause has as an "appositive." So they have used the term "supplementary relative" for a complement. I still think they are right when they call it "relative." I assume you differ there. It now seems to be a "relative-pro-article." She had long been accustomed to the solitary nature of her son's instincts, that I had tried--and failed-to stifle. I think it is clear from a careful reading of this sentence that the author did not intend that the first person be understood as trying to stifle all of her son's instincts. The author is talking about some of her son's instincts, namely those that the first person had tried to stifle. Paratactic paraphrase may help to display this. First the non-restrictive interpretation: She had long been accustomed to the solitary nature of her son's instincts I had tried--and failed-to stifle these instincts. Now the restrictive interpretation: I had tried--and failed-to stifle (certain ones of) her son's instincts. She had long been accustomed to the solitary nature of these instincts. My impression is that the difficulty arises when the strict interpretation of the noun phrase changes between the main clause and the dependent clause. Here it seems to be the specificity of the instincts that needs the complementation. Another vote for the "relative-pro-article." I hope these arguments don't go around in a circle again. I am convinced of the relative nature of that in adjective clauses, but its antecedent appears to be the normal domain of an article, rather than the full noun phrase. Bruce >>> [log in to unmask] 3/11/2005 8:27:47 AM >>> Bruce, I'm not sure yet that a three-way distinction works. The demonstrative we agree on. Here are the differences you outline between 2 and 3: 2 has only one morphological manifestation, occurs at the beginning of a content clause which may include appositives. 3 occurs at the beginning of an adjective clause, always restrictive, conjunction may be omitted Here are my problems. 3 also has only one morphological manifestation, so 2 and 3 are alike in that. 2 may be omitted, as in "I know it's late," so they're alike in that. Appositives (2) may start with "that". Huddleston&Pullum (p. 1052) note that, in what they call "supplementary relatives", their term for non-restrictives, "that" is allowed, as in February, that in other years held intimations of spring, this year prolonged the bitter weather. She had long been accustomed to the solitary nature of her son's insticnts, that I had tried--and failed-to stifle. 2 and 3 are thus alike in appositives may start with "that". That fact that one kind of that-clause identifies and another classifies is not a function of the "that" but of distinction between complement and modifier, which is why I'm uncomfortable with noun clauses after verbs, nouns, and adjectives as appositives. They're complements, not modifiers, and appositives are modifiers. Your claim that "(3)is typically an intergral part, yet as object can be omitted" begs the question. It assumes that (3) is a pronoun, but there is no evidence to support this claim. That-relatives in English, like relatives in a lot of languages, involve deletion of the embedded coreferent. In some complex cases a resumptive pronoun shows up, although English typically doesn't like resumptives. But consider Here's a book that I know the guy who wrote *0/it. Deletion would result in an island constraint violation. The resumptive, while inelegant, makes the sentence grammatical. The fact that a pronoun can appear in its in situ position is pretty strong evidence that "that" isn't the pronoun. The that of a that-relative is identical to the that of a noun clause. It's deletable, it's morphologically invariant, and it doesn't occur in places where a pronoun can, like after a preposition. 2 and 3 are indistinguishable. Herb Herb, Maybe Occam (Ockham) would object, but I don't have a problem with three "that"s. 1) The demonstrative. This appears before nouns. It also has a pronoun form for use without the noun. There are four distinct morphological instances of the demonstrative: this, these, that, those. 2) The conjunction for a noun clause. This is the variety of noun clause often called a content clause. There is only one morphological manifestation. The noun clause may appear in virtually all the functions of a noun. The appositive (adjectival) was one function under recent discussion. 3) The conjunction for an adjective clause. This is the variety of adjective clause used to identify rather than describe or classify. It is always restrictive. The conjunction is sometimes omitted in this kind of clause. The demonstrative appears as a pronoun. The idea of my last post is that you could argue that the conjunction of the noun clause (2) can be seen as being used as a pronoun as the connective for the adjective clause (3). The morphology of (2) and (3) is identical. The syntax is quite different: (2) is independent of the clause it introduces, whereas (3) is typically an intergral part, yet as object can be omitted. The fact that (3) functions in its clause as a part of it in the same way as a noun would do in an independent clause formulation seems to make it pronoun-like. Hence, the stretch that one might be justified in calling (3) a pronominalized version of (2). It is of some interest here with regard to pronunciation that the conjunction than was often indistinguishable from then in 17th Century England and is still so in many dialects. As conjunctives the first is a relative adverb of the degree clause while the second is either a relative adverb of a temporal clause (subordinate clause) or an illative conjunction (co-ordinate clause). The second form is also much like the demonstrative as an adverb. Thus we have three morpho-syntactic forms of similar syntactic distribution as that, but here they are adverbs. Hope this helps. Bruce >>> [log in to unmask] 3/10/2005 7:17:50 PM >>> Bruce, Interesting arguments! Jespersen treats comparatives similarly and also reports and describes the use of "as" to introduce a relative clause, as in cases like "not as many as I thought 0" and the less common "anybody as 0 wants to". You're right that my first three arguments apply more strongly to the question of whether relative "that" is the demonstrative pronoun. But consider the alternative. If my arguments don't apply to relative "that", then we have THREE, not two "thats", the conjunction that introduces noun clauses, the demonstrative pronoun, and the relative "that". But the relative "that" behaves morpho-syntactically like the conjunction, not like the demonstrative, so there's little reason for positing the third. "Pronominalizing subordinating conjunction"? I like that. Just think the potential it gives us as a model for naming lexical categories! Herb Herb, Perhaps your argument could be made stronger if you mentioned that certain adverb clauses are also relative, without having to be connected with a pronoun. In this case, however, someone might want to argue for a pro-adverb, so maybe it doesn't help after all. John is taller than George. John is taller than George is. John is taller than George is tall. The connective "than" refers to the degree of George's height, comparing John's height to it. It is relative by referring to the same thing that the -er is referring to, the degree of John's tallness. Your argument for "that" not being a pronoun seems only to show that it isn't the pronominalized demonstrative "that," at least in the first three points of the previous post. Perhaps we can view it as a "pronominalized subordinating conjunction." Bruce >>> [log in to unmask] 3/10/2005 12:41:42 PM >>> Craig, I don't think anyone questions whether wh-words are pronouns. That much is pretty clear. The problem is with "that". The morpho-syntactic evidence is overwhelming that relative "that" is not a pronoun and is a subordinating conjunction, that there is, in fact, no difference between the "thats" in I know that it's raining. and The rain that's falling now will flood the fields. They're the same thing. The claim that "that" in relative clauses is a pronoun is a claim grounded in a school grammar tradition that is seriously flawed in many ways, this being one of them. When you say "that is a pronoun in some camps and a complementizer in others when it functions within a relative clause," you beg the question. "That" in a relative clause has no function within the clause. It simply introduces it. It is not subject, object, OP, or anything else. Those relationships are marked by the absence of a noun phrase in the appropriate position, not by "that". Content clauses and relative clauses are similar in that they are both embedded sentences. They differ in that content clauses are complements of verbs, nouns, or adjectives and that relative clauses are modifiers of nouns. It is the modifier relationship that leads to the structural gaps exhibited by relative clauses but not by content clauses. I don't think the problem of appositives has anything to do with the analysis of "that". Rather, it has to do with the ill-defined nature of the term appositive. Here are some examples. 1. My brother Bill ... 2. Bill's statement that he was in Chicago at the time ... 3. Bill, who lives in Chicago, ... 4. Chicago, hog butcher to the world, ... 5. Bill's party, scheduled for last night, ... 6. The idea that Bill lives in Chicago ... etc. At best, appositive is a function, not a structure, and I'm not entirely sure that it's a function. I think rather that it's a traditional term used to describe a disparate variety of structures all of which occur after nouns. It has some usefulness if used with care. Calling 1,3,4,5 appositives doesn't bother me much, but including 2 and 6 does. I think they're different structures, complements to their head nouns rather than modifiers, and calling them appositives just confuses matters. But this is where poorly defined traditional grammar terms get us. Herb Herb, I know we have gone back and forth on this one before, and I'm still not convinced, but I think it may be important to clarify that there seems to be agreement that there is such a thing as a relative pronoun (who, with its various forms, and which, when functioning within these adjectival clauses), but that is a pronoun in some camps and a complementizer in others when it functions within a relative clause. We tend to agree that it is a complementizer in noun clauses precisely because it clearly has no role within the noun clause. I'm wondering whether you see any difference between a content clause structure and relative clause structure. (Are these the same structures, but differing in context by function?) The argument for these as appositional seems to hinge, at least for me, on the sense that that functions differently. Is the notion of appositional noun clause somewhat dependent on the misunderstanding of the role of that as pronoun, at least as you see it? Should we discard the category? Craig To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ ------------------------------------------------------------------------ ---- -- This message may contain confidential information, and is intended only for the use of the individual(s) to whom it is addressed. ------------------------------------------------------------------------ ---- -- To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ ------------------------------------------------------------------------ ---- -- This message may contain confidential information, and is intended only for the use of the individual(s) to whom it is addressed. ------------------------------------------------------------------------ ---- -- To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ ------------------------------------------------------------------------ ---- -- This message may contain confidential information, and is intended only for the use of the individual(s) to whom it is addressed. ------------------------------------------------------------------------ ---- -- To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ -- No virus found in this incoming message. Checked by AVG Anti-Virus. Version: 7.0.300 / Virus Database: 266.7.1 - Release Date: 3/9/05 -- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.0.300 / Virus Database: 266.7.1 - Release Date: 3/9/05 To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ ------------------------------------------------------------------------ ------ This message may contain confidential information, and is intended only for the use of the individual(s) to whom it is addressed. ------------------------------------------------------------------------ ------ To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/ To join or leave this LISTSERV list, please visit the list's web interface at: http://listserv.muohio.edu/archives/ateg.html and select "Join or leave the list" Visit ATEG's web site at http://ateg.org/