CB:>>But they changed qualitatively when they became alphabetic and no longer pictographic. There was a revolution in their "descent" when they shifted to alphabetic. They qualitatively shifted from iconic to arbitrary representation.<<
I doubt if they ever were truly iconic. In the shift to represent glottographic speech, the arbitrary quickly intrudes itself. Chinese characters are in effect as arbitrary as a syllabary or an alphabet. That is not to say, though, that any of these are ENTIRELY arbitrary. For example, quite unarbitrararily, English spelling points out word length (and at least hints at syllables), word breaks, and most importantly, word relationships. Not iconic, but linguistically motivated. English spelling balances phonetic/phonological with morphemic/lexical elements, making it look a lot like French. Or, if you will, a Germanic language in terms of pronunciation, looking like a Latin language. CB*>>If you mean the shape of the mouth on the "o", that is not the point of arbitrary representation. The arbitrary representation is not of the sound by the letter, but by the words made up of the letters in representing the things and concepts.<< I was point out <o> as motivated, that is, not entirely arbitrary. It is not by the arbitrary that letter <o> got chosen to represent sounds that involve lip rounding in English. On the other hand, it most likely represents the conscious choice of a Norman scribe or a Dutch typesetter somewhere in the history of the development of modern English spelling conventions. CB:>>Again, if you mean the letters picture the shape of the mouth in pronouncing them that is not the relevant representation. The arbitrary representation is in the spoken word itself representing things and concepts. So, that arbitrariness carries over to the alphabet of the writing. Whereas picture writing has come iconic, non-arbitrary representation of what is being represented as in Mixtec or Aztec codices. The actual Mixtec "king" or "queen" is pictured in the writing.<< I was pointing out non-arbitrary aspects of the ways in which roman alphabet has been adapted and/or evolved to represent English--glottographically. CB: >>How do you mean they are not discrete ? This is the crux of the whole idea. We _think_ of them as discrete when we hear them. Otherwise , we wouldn't be able to understand the word. We wouldn't be able to distinguish "pig" from "big", which we can.<< Not discrete, not digital but analog. Human speech is complexly analog. The pursuit of digital units has proven nice for formalization but largely irrelevant to applied linguistics, which must concern itself with things like language acquisition, use of languages across language and cultural groups, how to program a computer to process human speech, etc. CB:>>Aren't the syllables "pig" and "big" distinguished from each other by the phoneme at the beginning ?<< In terms of what speech perception research shows, they are differentiated by VOT--voice onset timing. If you still want to believe in phonemes, explain how a North American differentiates, for example, the pronunciation of 'ladder' from 'latter'. Also, remember, it is philosophically weak in any discussion to assume the very issue you are discussing to be true and existent--the phoneme exists, now let me show you how it exists. Rather, start with, does the unit of the phoneme (as defined by linguists) exist? Why is its existence not supported by current speech perception and articulatory phonetics research? CB>>CJ: As it has been identified and researched, categorical perception is not limited to human language or humans--even though it has been argued before as something that separates human language and humans from the rest of the animal kingdom. Most importantly though as it has been identified and researched, it appears to be an epiphenomenon.<<<<<< ^^^^ CB: What do you mean by epiphenomenon here ?<< Let 's go back to differentiating words. First of all, how do you differentiate, for example--to be quite arbitrary--pig from antidisestablishmentarianism? Second, how do you differentiate 'latter' from 'ladder' or 'utter' from 'udder', phonologically speaking. Third, if we can say we hear a [p-] at the beginning of 'pig' and ditto for a [b-] at the beginning of 'big', might such an act of identification, isolation, segmentation, etc. actually be a PRODUCT of successful word identification? In which case, the segmentation of the words and differentiation of the two supposed English phonemes, /p/ and /b/ are not causal to word identification but rather dependent on it--epiphenomenal to speech perception. To conclude with something different, here is a look at the non-arbitrary (that is, motivated) aspects of reduplication, which I had mentioned earlier. Some languages make considerably more use of this process, but it seems to be very basic to word creation in all languages. So again, not entirely arbitrary. http://www.icsi.berkeley.edu/ftp/global/pub/techreports/1994/tr-94-019.ps.gz Page 1 A Preliminary Study of the Semantics of Reduplication Terry Regier TR-94-019 April 1994 Abstract There is a universal component to the semantics of reduplication, which can be ex- pressed as a radial category of concepts. I present this radial category, along with supporting evidence from a range of languages, and motivations for the links between the various senses. The structure of the radial graph gives rise to a number of pre- dicted implicational universals. I also show that the radial category for reduplication shares an entire subsystem of concepts with the radial category for the Russian verbal prex raz- . This sharing of subsystems of concepts across separate radial categories suggests that there is a single universal core conceptual network, with individual constructions covering dierent, possibly overlapping, regions. The author may be reached by e-mail as [EMAIL PROTECTED] . Page 2 1 The Semantics of Reduplication Reduplication expresses a wide and initially somewhat arbitrary-seeming array of meanings cross-linguistically. There is, however, a core set of meanings which recur so frequently and in so many languages as to demand explanation. Consider for example the English words higgledy-piggledy, helter-skelter, pell-mell, and willy-nilly . These reduplicatives share a semantic component of lack of control with reduplicatives from many other languages, e.g. Russian tjap-ljap (in a slipshod manner) and Bengali ulat-palat (helter-skelter). Other senses which recur cross-linguistically include intensity, increased quantity, scattering, plurality, diversity, member of a class, repetition, continuation, more than appropriate, diminution,endearment,contempt, attenuation, augmentation, similarity, derogation, and habituality (Moravcsik 1978). In addition to these, Nomura & Kiy- omi (1993) note disorder, carelessness, pretense, reciprocity, aimlessness, and attempt. The apparent incoherence of the set of meanings she was examining led Moravcsik (1978:325) to conclude: Given that reduplication is neither the exclusive expression of any one meaning category in language, nor are the meanings that it is an expres- sion of all subsumable under general classes, no explanatory or predictive generalization about the meanings of reduplicative constructions can be proposed . [ Emphasis added ] In this work I hope to demonstrate that despite this bewildering diversity of mean- ings, some tentative explanatory and predictive generalizations can be made concern- ing the semantics of reduplication. I shall cast the various meanings expressed by reduplication as a radial category of concepts (Lako 1987). This construct allows one to express the relations between various senses of a linguistic unit, without as- suming that there exists a single abstract meaning which covers all these senses and no others. In general, the construction of a radial category involves the identication of a prototypical central sense and motivations for links radiating out from that sense to various other senses, and then possibly further links from those senses on to yet more peripheral senses. In using this construct to express universal tendencies in the semantics of reduplication, I adopt the notion of a universal radial category from Pederson (1991) and Jurafsky (1993): ... all languages which share the common prototype will have a set of extensions from that prototype which is a coherent subset of the universally- sanctioned potential extensional structure. That is, while each language may have a unique extensional structure, each language's extensional structure will consist only of nodes and links which are adjacent in the universal network. No language will have nodes which are non-adjacent in 1 Page 3 the universal network unless it also has (or has had) all the intermediate nodes as well. (Pederson 1991:2) Here the nodes represent individual senses or meanings, and the links represent mo- tivated conceptual relations between these meanings. The central prototype node is taken to represent not only the semantically central sense of the construction, but also the origin from which a history of semantic extension eventually led to the radial category as manifested synchronically (Pederson 1991; Nikiforidou 1991; Jurafsky 1993). This conception of a universal radial category lends some explanatory coherence to what might otherwise be a baingly heterogeneous set of meanings, and predicts implicational universals of the sort described above by Pederson. The use of radial categories in explicating the semantics of reduplication makes this work very similar in spirit to the work of Nomura & Kiyomi (1993), which uses radial categories for similar purposes. However, this work does not restrict its attention to verbal reduplication, as theirs did, nor to instances of reduplication in which the unreduplicated stem is meaningful. On the other hand, their coverage of the world's language families is more thorough. We shall see below that the radial category for reduplication shares an entire subsystem of concepts with the radial category for the Russian verbal prex raz- . In other words, many of the same extended senses are expressed both by reduplication and by raz- , although reduplication and raz- have dierent central senses. This sharing of conceptual structure suggests the existence of a single interconnected conceptual network, with individual linguistic constructions rooted at dierent nodes, covering possibly overlapping subsets of nodes. Figure 1 illustrates some of the various senses of reduplication, and their inter- connections. The senses shown in ellipses are shared with raz- , while those shown in rectangles are not. The central sense, repetition, is shown in a double rectangle. The graph structure shown here divides quite cleanly into two halves: the subgraph on the left, which is not shared with raz- , and that on the right, almost all of which is. The links between senses are directed, indicating the direction in which I am proposing semantic extension took place. Traugott (1982) has argued for the unidirectionality of semantic change, in general from a concept concerning some state or object de- scription to a concept concerning the speaker's attitude. As we shall see, some of the links shown here represent extensions of this sort, e.g. the link from small to contempt. Evidence for the senses shown here, and motivation for the links between them, are presented below. Whenever possible, independent motivation for the various links is adduced. As this work is preliminary in nature, not all of the senses found by Moravcsik (1978) and Nomura & Kiyomi (1993) are covered here, although some new ones are, e.g. insect and bird. In addition, it may be the case that some of the particulars of the graph structure presented here are incorrect or incomplete. A more accurate model will have to await the incorporation of more data, hopefully including 2 Page 4 = a sense shared with raz- = a sense not shared with raz- non-uniformity intensity completion bird contempt aection incrementality continuity insect lack of specicity plurality lack of control repetition scatter spread out, baby small Figure 1: Radial category for the semantics of reduplication. 3 Page 5 historical data indicating the course of semantic extension of reduplication. Repetition English boogie-woogie Mongolian bayn bayn often, constantly Sundanese guguyon to jest repeatedly (guyon - to jest) Tzeltal -pikpik to touch it lightly repeatedly (-pik - to touch it lightly) I take repetition as the central sense of reduplication. In choosing this as the central sense, I dier from Nomura & Kiyomi (1993), who take an abstract notion of replication as the center, which then gives rise to repetition (and continuation) as \replication across time", and plurality as \replication across participants". I have chosen repetition as the center primarily because it, unlike either plurality or the more abstract replication, is directly iconically grounded in the repetition of the stem in the linguistic form. Plurality The use of reduplication to express plurality is widespread, appearing in Dakota, Agta (Niepokuj 1991), Comox (Sapir 1915), the Australian language Djaru (Nomura & Kiyomi 1993), Papago, Samoan, and numerous other languages (Moravcsik 1978). There are several variations on this general theme, reduplication sometimesexpressing simple nominal plurality, sometimes plurality of verbal subject, sometimes plurality of verbal object, etc. Plurality is closely related to the central sense, repetition, in that the repetition of the stem in reduplication results directly in plurality of the number of times the stem has been uttered. Incrementality Farsi kam kam gradually Mongolian tus tus each separately, each individually (tus - this, this same) Pal. Arabic swaysway little by little (sway - a little) Tangale dodok-dodok one by one (dodok - one) Incrementality is related to repetition through a PART FOR WHOLE metonymy, a conceptual link used widely throughout language. An example of this is the En- glish expression fteen head , e.g. of cattle, in which the word head , a part of the individual cow, is used metonymically to refer to the cow as a whole. This metonymy is applicable here since incrementality, the notion of performing some action bit by bit, involves the repeated application of some subpart of the overall action, gradually 4 Page 6 completing the action as a whole. Repetition is therefore a part of the semantics of incrementality, so the use of a form meaning repetition to also mean incrementality is well motivated metonymically. Continuity The use of reduplication to express continuity has been noticed by many researchers. Reduplication expresses continuative aspect in Tagalog (French 1988) and Javanese (Niepokuj 1991). In addition, the continuative sense of reduplication is evident in Iraqi Arabic t . ant . an (to buzz or hum), Hindi kit kit (monotonous droning on), and arguably English dilly-dally . The use of reduplication in the sense of continuity is easily motivated from repetition using the well-established MULTIPLICITY TO MASS image schema transformation (Lako 1987:428). This transformation appears in many other aspects of language, such as the polysemy of English over : one can say, for example, The guards were posted all over the hill , implicitly viewing the multiple guards as a mass which covers the hill. While this transformation is often used in reference to physical mass and multiplicity, it is here analogously used in the temporal domain: continuity is the mass form of repetition. Insect Azeri mig . mig . a mosquito Comox xwaxwadjom y Syr. Neoaramaic 'ir'ra wasp A number of languages have reduplicative names for buzzing insects. An intu- itively appealing motivation for this is simply that these insects make continuous sounds, and this association provides a conceptual link from continuity to insect. We shall see a number of links below which are similarly motivated. Bird Atsugewi qasqasa crow Comox xopxop humming-bird English cuckoo Spanish caracara Tamil kaka crow It is quite common for birds to have reduplicative names. The motivation for this is entirely analogous to the motivation for insects: birds often make repetitive sounds, providing a simple conceptual link from repetition to bird. 5 Page 7 Baby English baby French bebe baby Tamil papa baby A number of languages use reduplicative nouns for the concept baby. This is presumably the case because babies, like birds, often make repetitive sounds, yielding an associational link from repetition to baby. In addition to this, and related to it, reduplication can be used to mark \baby register", that register used when addressing babies. Consider for example English teeny-weeny, Georgie-Porgie , and Farsi rizeh-mizeh (teeny-weeny, rizeh - little). There is also the related use of reduplication to express aection, noted by Moravcsik (1978) and exemplied by Palestinian Arabic Dandun , the hypocoristic form of the name Dina. This extension from baby to aection is the sort of unidirectional semantic spread for which Traugott argues, from concrete reference to an expression of the speaker's attitude. Small Comox djidjidis little tooth (djidis - tooth) English tidbit English nitty-gritty Farsi rizeh-mizeh teeny-weeny (rizeh - little) The use of reduplication to express diminution, i.e. the concept small, is fairly widespread, and is noted by both Moravcsik (1978) and Niepokuj (1991). The link from baby to small is motivated by the perceptually very salient fact that babies are small, and may well be the most conceptually salient class of small things. The use of linguistic forms meaning baby to also mean small is illustrated in such English sentences as Look, a baby airplane , meaning a small airplane. Contempt Bengali bhethor-shethor in, but that is insignicant (bhethor - in) Dutch mik-mak worthless collection English claptrap English hillbilly German Pille-palle insignicant things Yiddish layfen-shmayfen running is beside the point (layfen - to run) A number of languages use reduplication to express contempt, insignicance, or irrelevance. This can be motivated via the metaphorical conceptual link POWER 6 Page 8 AND IMPORTANCE IS SIZE, linking small to contempt (Jurafsky 1993). Jurafsky, in his work on the semantics of the diminutive, points out that this metaphor can be seen in the semantics of the English -y sux: while often used for diminution (e.g. doggy ), -y can also express derogation or contempt (e.g. limey , a derogatory term for an Englishman). This extension, like the extension from baby to aection, is an instance of directed semantic spread from the concretely referential to the attitudinal. Lack of Control Bengali ulat-palat helter-skelter Danish misk-mask mess, disorganized jumble Dutch schelle-belle overly independent young woman English helter-skelter Farsi g . = ti-p = ti pell-mell German Remmidemmi noisy event, e.g. party Pal. Arabic harj w marj confusion, hubbub Russian tjap-ljap anyhow, in a slipshod manner Lack of control and disorder are expressed by reduplication in a number of lan- guages; in addition to the ones shown above, Nomura & Kiyomi (1993) cite examples from the Bantu languages Southern Sotho, Luba-Shaba, and Ndengese. A plausible motivation for this phenomenon springs from the fact that babies are often out of control, doing things we wish they wouldn't. This gives us a simple associative link from baby to lack of control. Spread Out / Scatter Japanese tokorodokoro scattered Mongolian aravgar-saravgar spread out Tamil inge-ange here and there Uzbek alang-jalang looking around The notion of scattering or spreading out is also often expressed by reduplication. Motivation for this can be found in the fact that lack of control can cause scattering or spreading out. The image of an out-of-control child throwing food around may be helpful in making this association concrete. This link is also highlighted in the colloquial American English sentence He's a very together person , which uses a word whose central sense is the opposite of dispersion or distension to express self-control, i.e. the opposite of lack of control. The English word scatterbrained makes use of the same mapping. In these English examples the extension operates in the other direction, in that they use lexemes indicating scattering (or lack thereof) to denote control (or lack thereof), while in reduplication the directionality is reversed. They do provide independent motivation for the existence of the link, however. 7 Page 9 There is an interesting phenomenon in Palestinian Arabic related to this sense of reduplication: many small sprinklable and hence scatterable foods have reduplicative names, e.g. pepper ( ll ), sesame ( simsim ), mint ( na'na' ), and crumbs ( fataf it ). While apricot ( mismis ) does not seem to t the general pattern at rst, it may be signicant that in the Levant apricot is often beaten out at into sheets, dried, and sold in that form. This gives us an obvious connection to the sense of spreading. Lack of Specicity Farsi hulu-mulu peaches and such (hulu - peach) Latin quisquis whoever (quis - who) Pal. Arabic kaza-kaza whatchamacallit Sundanese sahasaha whoever (saha - who) Lack of specicity is expressed by reduplication in a range of languages, including those shown above, Khasi and Turkish (Moravcsik 1978), Hindi, and Uzbek. The motivation for this comes from lack of control, in that these examples can be seen as denoting lack of control in referential specicity, a sort of sloppy uncontrolled scatter- ing motion in semantic space, picking out a cluster of closely related meanings rather than just a single one. There appears to be an interesting areal phenomenon encom- passing Turkish, Uzbek, Farsi, and Hindi, in which the use of partial reduplication with the initial consonant of the stem altered to a labial in the second occurrence de- notes lack of specicity. Consider for example Turkish kitap-mitap (books and such, kitap - book), Uzbek n = n-p = n (bread and such, n = n - bread), Farsi hulu-mulu (peaches and such, hulu - peach), and Hindi kitab-vitab (books and such, kitab - book). Non-Uniformity English hodge-podge German zick-zack zig-zag Kurd. Neoaramaic range-range in several colors (range - color) Tamil ippo-appo now and then Non-uniformity is sometimes expressed by reduplication, as shown by the exam- ples above. The primary motivation for this is the metaphor STATES ARE LOCA- TIONS, which transforms physical scatteredness, or non-uniformity of location, to non-uniformity of state, taking us from scatter to non-uniformity. This widely-used metaphor can also be seen in such expressions as They're moving toward reconcilia- tion , in which progress toward the abstract social state of reconciliation is expressed in terms of physical motion towards a goal. In addition to this motivation, one could imagine a link from plurality to non-uniformity, in that non-uniformity implies a plurality of dierent states. 8 Page 10 It is interesting to note that the sense of non-uniformity is often expressed by non-uniformity of form, i.e. by partial rather than complete reduplication. Many English reduplicatives with this sense seem to exhibit variance in form between the rst and second instantiations of the stem e.g. mish-mash, hodge-podge, knick-knacks, bric-a-brac, zig-zag . This need not be the case in all languages of course; Neoaramaic range-range is a counterexample. Intensity English a whole whole lot Hindi lal-lal very red (lal - red) Mongolian araycaray just barely (aray - barely) Mongolian ocnoon tocnoon a great deal Reduplication is often used to express intensity. In addition to the examples listed above, this sense appears in Dakota, Turkish, Dagur, Perak Malay, Tangale (Niepokuj 1991), Sundanese, and Thai (Moravcsik 1978). There are at least two possible sources of motivation for this use. One of these is a link from plurality, highlighted by the English expressions many thanks (Moravcsik 1978) and a thousand pardons , in which quantity is used to express intensity. The other is a link from spread out, or perhaps more accurately from the result of spreading out, from enlargement. For example, we can see from the Russian expression bol'shoe spasibo (literally, big thank-you) that size can also be used to express intensity. Completion Completion is also sometimes expressed by reduplication. An example of this from the Bantu language Ewondo is b = -b == (to make/do completely), where b = means to make or do (Nomura & Kiyomi 1993). In addition, Moravcsik (1978) notes that reduplication marks the perfect in Sanskrit, Greek, Latin, and Gothic. A plausible motivation for the use of reduplication in this sense stems from intensity: performing an action intensely can lead to completion of the overall action. For example, working intensely on a research project will tend to result in completion of the project. Discussion There are a number of senses of reduplication I have not incorporated into the radial category presented here. Some are extremely rare, and are probably best viewed as aberrations, e.g. the use of nominal reduplication to express third person singular possession in Tarok (Niepokuj 1991:10). Others, such as the concept game are ap- parently fairly widespread: consider English ping-pong, pall-mall, tic-tac-toe , Basque (and now Spanish and English) jai-alai , Hindi holi-koli , and ses-bes , a Levantine vari- 9 Page 11 ety of backgammon. 1 Senses such as this one, and the numerous others already noted by other researchers, are common enough that their inclusion in the radial category would be required for a more complete account of the semantics of reduplication. Another serious aw in the work as it currently stands is that its coverage of the world's languages is quite limited. Despite these and no doubt a number of other shortcomings, I hope that this analysis provides some explanatory insight into the polysemous structure of redupli- cation, and helps to make sense of an initially rather random-seeming welter of data. In many cases, the conceptual links between senses can be independently motivated by non-reduplicative data, and in some cases they are in keeping with the previously observed tendency of unidirectional semantic extension from the concrete to the at- titudinal. Perhaps the most satisfying aspect of a model of this sort, however, is that it makes concrete and falsiable predictions. Following Pederson (1991), if we view the universal radial category for reduplication as not only an indication of universally sanctioned synchronic structure, but also one of historical semantic development, pre- dictions arise from the structure of the graph itself. Specically, if we view the graph as a trace of semantic extension over time, we would predict that any language in which reduplication expresses some peripheral sense would also have, or at some time have had, reduplicative terms expressing the central sense of repetition, along with all intervening senses on a path from the center to the peripheral node in question. For example, we would predict that if a language uses reduplication to express spread out, it would also use, or have used, reduplication to express repetition, baby (presumably including baby register), and lack of control, as these concepts lie on the only path from the central sense repetition to spread out. While the work is still somewhat speculative in nature, the process of trying to falsify the predictions generated by the model as it stands should prove helpful in constructing a more accurate model. 2 The Semantics of the Russian Verbal Prex raz- The radial category for reduplication shares a number of interconnected concepts with the radial category for the Russian verbal prex raz- . This can be seen from Figure 2, which illustrates some of the more frequently attested senses of raz- , and the links between them. We have already seen most of these from our look at the semantics of reduplication. In particular, the senses shown in ellipses are shared with reduplication, while those shown in rectangles are not. Roughly the same radial 1 This last example is particularly intriguing. After throwing the dice it is customary to call out the numbers that appear, and in the Arabophone Levant this is done in Farsi for almost all pairs of numbers. A notable exception is the pair six-ve. In this case, the six is called out in Farsi ( ses ) but the ve is called out in Turkish ( bes ), and the game is named after this combination. It is at least possible that this departure from the use of Farsi alone is the result of coercing the name of the game into a reduplicative template, done presumably because of some as-yet-undetermined conceptual link indicating that games are appropriately named using reduplication. 10 Page 12 intensity completion = a sense not shared with reduplication = a sense shared with reduplication non-uniformity spread out, scatter lack of control analysis split Figure 2: Radial category for the Russian verbal prex raz- . 11 Page 13 category of senses seems to hold for the use of raz- in some other Slavic languages. Intriguingly, Atanasova et al. (1992) list \harum-scarum" as the meaning of Bulgarian razvej-prax , highlighting the semantic overlap between raz- and reduplication with which we are concerned here. This sharing of structure is signicant since it suggests that the concepts and the links between them are not peculiar to reduplication, or to raz- , but are rather part of a universal core conceptual network, subsections of which may be expressed by various linguistic constructs. The idea then is that the radial categories for reduplication and raz- simply give us snapshots of subregions within a much larger conceptual network, and the radial category overlap gives us an indication of how the snapshots t together. I take the central sense of raz- to be spread out, scatter. From here, there are links to the non-shared senses of split and analysis, as well as conceptual links we have al- ready discussed leading to intensity, completion, non-uniformity, and lack of control. In all cases except one, these links are of the same directionality as the corresponding links we saw in the radial category for reduplication. The exception is the link from spread out, scatter to lack of control. I take the fact that the same link appears with two directionalities in the two radial graphs as evidence of an underlying bidirection- ally sanctioned link between the two concepts, that is, a link in the overall conceptual network which will allow semantic extension to travel across it in either direction. It is critical to draw a distinction between the directionality of semantic extension from one concept to another as posited for a particular linguistic form such as reduplica- tion , and the directionality which that link in the overall conceptual network may assume in general. Traugott's work suggests that some links will be unidirectional not only from the point of view of radial categories for individual linguistic forms, but in the overall conceptual network generally. Others however, such as the one in question here, may be potentially bidirectional. I present below a number of examples of the use of raz- in its various senses. Since all links except the ones from spread out to split and from split to analysis have already been discussed, I shall motivate only these two. Spread Out / Scatter razgonjat' to disperse razmetat' to scatter (s.t.); to spread (s.t.) out raznosit'sja to spread raskidyvat' to scatter; spread rasprostranjat' to spread, distribute rasseivat' to disperse, scatter rassredotochivat' to disperse rasstilat' to spread 12 Page 14 Lack of Control razbushevat'sja to rage, get violent, start lashing out razvolnovat' to upset (s.o.) razlazhivat'sja to go wrong razmechtat'sja to be lost in dreams razozlit'sja to get furious razrydat'sja to burst into sobs raskapriznichat'sja to become very naughty, act up rasserdit' to annoy, to make angry Split razgorazhivat' to partition (s.t.) razgryzat' to bite (s.t.) in two razdvaivat' to divide (s.t.) into two, split in two razdeljat'sja to break up, split up razluchat' to separate raznimat' to part; to separate Split can be motivated by viewing it as a special case of scatter: when the number of objects to be scattered is two, to scatter is to split or separate. Analysis razgljadyvat' to examine razdumyvat' to think, ponder razlichat' to make out, discern; to distinguish between rassledovat' to investigate, inquire into rassmatrivat' to consider, examine rassuzhdat' to reason rasschityvat' to calculate These words appear to have in common a semantic elementof analysis. This can be motivated from split through the metaphor MENTAL OBJECTS ARE PHYSICAL OBJECTS, evidenced in the English phrase tossing a few ideas around . To analyze, then, is to separate and be able to distinguish between mental objects, much as to split is to separate physical objects. 13 Page 15 Non-Uniformity razdumat' to change one's mind razlichat'sja to dier, be distinguished raznit'sja to dier raznoobrazit' to diversify raskhodit'sja to disagree, dier Intensity razbalivat'sja to be or become properly ill razobidet'sja to take great oense (obidet'sja - to be oended) raskaljat' to make scorching hot rasshalit'sja to get very playful Completion raskupat' to buy up raspivat' to empty a bottle drinking razgadyvat' to solve; to get to the bottom of razljubit' to stop loving razrjazhat'sja to run down, be used up razygryvat' to bring to a conclusion In addition to the examples of completion shown above, raz- can be used as a perfective marker; e.g. rasserdit'sja is the perfective form of serdit'sja (to become angry). It is presumably noncoincidental that this verb has a semantics involving lack of control, another sense of raz- . I have not yet determined whether all Russian verbs which take raz- as a perfective marker have a base semantics implicating one of the senses listed above. 3 Conclusions The striking cross-linguistic regularities which appear in the semantics of redupli- cation can be explained through the construction of a universal radial category of concepts, following Pederson (1991). This radial category may be interpreted as in- dicating not only synchronic relations between the various senses of reduplication, but also a possible historical trace of semantic extension from the central sense of repetition out to senses whose motivation is less obvious. Importantly, it is not the case that each language must have each sense shown in the graph; however, the graph does predict which senses enable extension to which others. Thus, the model will pre- 14 Page 16 dict that any language which uses reduplication to express a peripheral sense will also use reduplication (or have used it at some time in its past) to express all senses on a path from the central sense to the sense in question. In general, this work will have been successful to the extent that it lends some explanatory coherence to what would otherwise be a rather haphazard-seeming set of cross-linguistic regularities. Interestingly, the semantics of reduplication and the semantics of the Russian verbal prex raz- overlap to a surprising degree: an entire subsystem of concepts and links is shared. The signicance of this shared conceptual structure is that it suggests the existence of a single unied conceptual network, such that the the radial categories for reduplication and raz- are simply overlapping subregions in this larger network. This study has been exploratory and preliminary in nature; much remains to be done. To buttress claims of universality, more languages must be examined, from dierent language families. Currently unattached senses of reduplication such as game must be incorporated into the radial category. And perhaps most importantly, the predictions implicitly made by the graph structure of the universal radial category for reduplication must be empirically tested. I leave these tasks for future research. Acknowledgements Many thanks to my informants: Dina Aburous (Palestinian Arabic), Lone Hansen (Danish), Dirk Geeraerts (Dutch), Nasir Kanbarov (Uzbek), Naghmeh Mirghafori (Farsi), Srini Narayanan (Bengali, Hindi, Tamil), Michael Rehse (German), Lokendra Shastri (Hindi), Andreas Stolcke (German), and Len Talmy (Atsugewi, Yiddish). Thanks also to Dan Jurafsky, Srini Narayanan, Clay Scott, Len Talmy, and Nancy Urban for enjoyable and helpful conversations related to this work. Tzeltal data from Berlin (1963), seen in (Moravcsik 1978); Tangale data from Kidda (1985), seen in (Niepokuj 1991); Sundanese data from Robins (1959); Az- eri data from Haneda & Ganjelu (1979); French data from Mansion (1991); Iraqi Arabic data from Erwin (1963); Mongolian data from Bosson (1964); Japanese data from Matsuda (1974); Kurdish Neoaramaic data from Krotko (1982); Latin data from Glare (1983); Syrian Neoaramaic data from Spitaler (1938); Turkish data from Moravcsik (1978); Comox data from Sapir (1915); Russian data from Taube et al. (1987). References Atanasova, T. , M. Rankova , & V. Filipov . 1992. Bulgarian-English Dictionary . Soa: Izdatelska Kushta VEZNI-4. 15 Page 17 Berlin, Brent . 1963. Some semantic features of reduplication in Tzeltal. Interna- tional Journal of American Linguistics 29.211{218. Bosson, James E. 1964. Modern Mongolian { A Primer and Reader . Bloomington, Indiana: Indiana University Publications. Uralic and Altaic Series, volume 38. Erwin, Wallace M. 1963. A Short Reference Grammar of Iraqi Arabic . Washing- ton DC: Georgetown University Press. Arabic Series, number 4. French, Koleen Matsuda . 1988. Insights into Tagalog: Reduplication, Inxation, and Stress from Nonlinear Phonology . Arlington, Texas: Summer Institute for Linguistics and University of Texas at Arlington. Glare, P. G. W. 1983. Oxford Latin Dictionary . New York: Oxford University Press. Haneda, Koichi , & Ali Ganjelu . 1979. Tabrizi Vocabulary: An Azeri-Turkish Dialect in Iran . Tokyo: Institute for the Study of Languages and Cultures of Asia and Africa. Jurafsky, Daniel . 1993. Universals in the semantics of the diminutive. In Pro- ceedings of the Nineteenth Annual Meeting of the Berkeley Linguistics Society . University of California, Berkeley. Kidda, M. , 1985. Tangale phonology: a descriptive analysis. Ph.D. dissertation, University of Illinois. Krotkoff, Georg . 1982. A Neo-Aramaic Dialect of Kurdistan: Texts, Grammar, and Vocabulary . New Haven, Connecticut: American Oriental Society. Lakoff, George . 1987. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind . University of Chicago Press. Mansion, J. E. 1991. Harrap's Standard French and English Dictionary . London: Harrap Books. Matsuda, Koh . 1974. Kenkyusha's Japanese-English Dictionary . Tokyo: Kenkyusha. Moravcsik, Edith . 1978. Reduplicative constructions. In Universals of Human Language , ed. by Joseph H. Greenberg, 297{334. Stanford University Press, Stan- ford, California. Niepokuj, Mary Katherine , 1991. The historical development of reduplication, with special reference to Indo-European. Ph.D. dissertation, University of Cali- fornia, Berkeley. 16 Page 18 Nikiforidou, Kiki . 1991. The meanings of the genitive: a case study in semantic structure and semantic change. Cognitive Linguistics 2.149{205. Nomura, Masuhiro , & Setsuko Kiyomi , 1993. How to motivate the meanings of verbal reduplication: Cognitive and typological perspectives. Handout at Berke- ley/UCSD cognitive linguistics workshop, University of California, San Diego. Pederson, Eric William , 1991. Subtle semantics: universals in the polysemy of reexiveand causative constructions. Ph.D. dissertation, Universityof California, Berkeley. Robins, R. H. 1959. Nominal and verbal derivation in Sundanese. Lingua 8.337{369. Sapir, Edward . 1915. Noun Reduplication in Comox, a Salish Language of Vancou- ver Island . Ottowa, Canada: Government Printing Bureau. Canada Department of Mines, Geological Survey, Anthropological Series, number 6, memoir 63. Spitaler, Anton . 1938. Grammatik des Neuaramaischen Dialekts von Ma'lula (Antilibanon) . Leipzig: Deutsche Morgenlandische Gesellschaft. Taube, A. M. , I. W. Litvinova , A. D. Miller , & R. C. Daglish . 1987. Russian-English Dictionary . Moscow: Russky Yazyk Publishers. Traugott, Elizabeth Closs . 1982. From propositional to textual and expressive meanings: some semantic-pragmatic aspects of grammaticalization. In Perspec- tives on Historical Linguistics , ed. by Winfred P. Lehmann & Yakov Malkiel, 245{271, Amsterdam. Benjamins. 17 _______________________________________________ Marxism-Thaxis mailing list [email protected] To change your options or unsubscribe go to: http://lists.econ.utah.edu/mailman/listinfo/marxism-thaxis
