COMPOUNDS IN BRAZILIAN SIGN LANGUAGE: DEFINING WORD FORMATION PATTERNS

COMPOUNDS IN BRAZILIAN SIGN LANGUAGE: DEFINING WORD FORMATION PATTERNS Aline Garcia Rodero Takahira1 – USP – [email protected] Rafael Dias Minussi2 – USP...
2 downloads 0 Views 329KB Size
COMPOUNDS IN BRAZILIAN SIGN LANGUAGE: DEFINING WORD FORMATION PATTERNS Aline Garcia Rodero Takahira1 – USP – [email protected] Rafael Dias Minussi2 – USP – [email protected] Abstract: This paper aims at investigating word formation strategies in Brazilian Sign Language (LSB, from Portuguese, Língua de Sinais Brasileira), and search for a definition of what a morpheme in this language is. Its specific objective is to explore compounding, a strategy often used in word formation, studying its morphosyntactic behavior, besides defining general patterns in compounds in LSB. This paper follows Distributed Morphology (DM) (Halle & Marantz, 1993; Marantz, 1997) theory. We are considering both non-simultaneous compounds as CASA^ESTUDAR (escola) / HOUSE^TO-STUDY (school) and the possibility of forming simultaneous compounds as SEXO^VIAJAR (lua-de-mel) / SEX^TO-TRAVEL (honeymoon). Taking examples as the ones shown in Figueiredo Silva & Sell (2009) and others collected by recorded data and by interviews with native signers of LSB, mainly in the variation used in São Paulo City, we raised some hypothesis: (i) there are two kinds of roots in LSB, one in which the feature of movement is incorporated and other in which the feature of movement is not incorporated; and (ii) the feature [+movement] can be considered a categorizing morpheme, when it is not incorporated into the root to form its meaning, in a way that this hypothesis follows from the first. As results for this research, we examined the kinds of productive compounds and identified some properties present in the compounds of LSB, as the reduction or lack of the feature of movement in most part of compounds, suggesting that compounds in LSB are not formed from verbs, but from categorical neutral roots and classifiers (in sign language sense – see footnote number 7). Keywords: Brazilian Sign Language; Compounding; Distributed Morphology, Word formation.

1. Introduction3 In this paper we are discussing compounding patterns in Brazilian Sign Language (LSB, from Portuguese, Língua de Sinais Brasileira, or LIBRAS4). Since there are few papers about morphology in LSB, as Brito (1995), Quadros & Karnopp (2004), Felipe (1998 5 and 2006), Figueiredo Silva & Sell (2009) and Rodero-Takahira (2012), we have also appealed for papers about morphology in American Sign Language (ASL), as Stokoe (1960), Lidell & Johnson (1986), Aronoff, Meir & Sandler (2005), Del Giudice (2007), and Meir et al. (2010), among others. We investigated some criteria proposed by Felipe (2006) that takes as morphemes: i) ii) iii)

a root or a stem, considering Movement; an affix, considering alternations between Movement and Handshape; or, a desinence, as direction – considering number-person agreement; or Handshape – considering gender6.

1

Phd student at USP. Advisor: Prof. Ana Paula Scher. I thank CAPES for the financial support. Phd in Linguistics at USP. 3 We would like to thank two anonymous readers who commented on the first versions of this paper, and Rodrigo Florentino da Silva, our Deaf informant who provided some data for us. 4 LSB is the international form used to refer to Brazilian Sign Language. However, in Brazil, it is best known as LIBRAS. In this paper we decided for using the international form, LSB. 5 We didn’t have access to this paper once it has not been available at the library at UFRJ any longer. 6 In that paper, “gender” refers to animated – person, animal and inanimated – thing, vehicle. 2

1

We wonder if the parameter Movement, a phoneme as is explained in literature about sign languages, can really be considered a root in all kinds of signs, even in the ones that are signed only by Facial Expressions, as LADRÃO7 (“thief”). The specific objective of this paper is to explore compounding, studying its morphosyntactic behavior, besides defining general patterns in word formation in LSB. We are considering compounds as in (1): (1)

a. CASA^ESTUDAR (escola) HOUSE^TO-STUDY (school) b. ÁGUA^CAMINHO (rio) WATER^PATH (river) c. LOJACL^PIZZA (pizzaria)8 STORECL^PIZZA (pizzeria) d. SEXO^VIAJAR (lua-de-mel) SEX^TO-TRAVEL (honeymoon) e. LADRÃO^ATRAIR (“ganhar o público”) THIEF^TO-ATTRACT (“to win audience”) f. PÉ^PÉCL (chinelo) FOOT^FOOTCL (flip flop)

the

The main question discussed in this paper, in relation to compound formation as the ones in (1), is related to the fact that authors such as Felipe (2006) consider that there is no difference between a noun and a verb in some pair in LSB since the deaf they interviewed couldn’t realize that difference. On this way, we know neither whether a compound in LSB is formed with nouns or with verbs, nor we know exactly what can be considered a morpheme on this language. Data as the ones in (2), then, would not present differences, according to Felipe (2006); (2)

a. CASA / MORAR b. TELEFONE / TELEFONAR c. CADEIRA / SENTAR d. TESOURA / CORTAR COM TESOURA WITH-A-PAIR-OF-SCISSORS)

(HOUSE / TO-LIVE) (TELEPHONE / TO-CALL) (CHAIR / TO-SIT-DOWN) (PAIR-OF-SCISSORS / TO-CUT-

Taking the examples as those shown in Figueiredo Silva & Sell (2009) and others collected by recorded data and by talks with native signers of LSB, mainly in the variation used in São Paulo City, we raised some other questions: 1- What is a morpheme in LSB?; 2- What is a root or stem? 3- What kinds of compounds does LSB admit?; and; 4- What are the structures of the compounds in LSB?

7

The transcription system used in this paper uses words from Brazilian Portuguese in capital letters to indicate the signs; hyphen to indicate when more than one word in Portuguese is used to describe a single sign/; ^ to indicate compounds; and, @ in the place of gender inflexion to show the absence of this morphological mark in LSB. This system is the same adopted in Figueiredo Silva & Sell (2009) and Quadros & Karnopp (2004), among others. As a matter of space, we include neither pictures/draws in this paper to illustrate the signs, nor the form for written LSB (SignWriting). The pictures and signwriting form can be seen in the LSB dictionary (CAPOVILLA & RAPHAEL, 2001). For more pieces of information about transcription systems, also see McCleary & Viotti (2007). 8 CL stands for classifier. In sign languages classifier is a handshape used to show how something or someone is. For example, the verb “walk” is going to be different if it is a person, a dog or a bird walking, what changes is the classifier used to shoe “legs of a person shape”, “feet of a dog shape” or “feet of a bird shape”. In this paper we are going to use both the notion of classifier in sign languages and the notion of classifiers in oral languages, which we will see later in this paper.

2

This paper follows Distributed Morphology (DM) (Halle & Marantz, 1993; Marantz, 1997) and papers about compounds within this theory, as Minussi (2009). This theory assumes that both words and sentences are formed during the syntactic derivation, in this way, there is no lexicon that previously forms words. Instead, DM predicts that there is a list with primitive elements that contains acategorial roots and abstract morphemes. For this reason, this theory may bring a new point of view about the content of words in LSB and helps us to define the category of words, as well as to investigate the hypothesis that what is behind the formation of compounds in LSB seems to involve an analysis of the roots that form the compounds as the ones presented. DM poses the existence of three Lists, as we can observe in the architecture of grammar proposed by the model (Figure 1).

Figure 1 Architecture of DM (SIDDIQI, 2009, p. 14)

In the present paper, we are going to focus on the content f List 1 (List of morphosyntatic features). In oral languages, this List contains roots and abstract features as [+past], [+def], 3rd person, [+PL] (cf. HALE & MARANTZ, 1997; SIDDIQI, 2009). In some versions of this theory, as in Arad (2005) and Minussi (2008, 2009) this List contains acategorial roots, that are a layer of semantic abstract features that form a general concept. List 1 is the List that will feed the syntactic component for the words and sentences formation. Since there are no papers in LSB that uses DM, it is necessary to define what we are treating as a root in LSB, besides verifying which features that seem to be relevant for word formation in this language. This paper is divided in the following way: in section 2, we discuss which features form a root in LSB, following what Marantz (2001) and Arad (2003) consider as a root, and we also discuss that there is a difference involving movement between pairs of nouns and verbs for some of the deaf, based on our recordings and data presented in Felipe (2006); in section 3, we show the compounding types in LSB shown in literature as in Felipe (2006) that discusses different examples of juxtaposition in LSB and Figueiredo-Silva & Sell (2010) that 3

consider some rules, as fixed order, to identify what they call “true compounds”, and we also hypothesizes two structures of compounds in LSB: [√ROOT (neuter) + classifier] and [√ROOT (neuter) + √ROOT (neuter)]; in section 4, we discuss the possibility of forming simultaneous compounds in LSB, based on Meir et al. (2010) about ASL; finally, in section 5, we bring the conclusions of this paper. 2. About the features that a root contain In this section, we are going to discuss in details how a sign may be formed in LSB, taking into account that we are adopting DM theory, as we have mentioned above. DM discusses that, for oral languages, both words and sentences are formed by the syntactic component. For sign languages, as we have already said, it is necessary to make some theoretic adjustments, as for example, what we are in fact considering as a root and what the abstract morphemes that we find in these languages are, besides the fact that the vocabulary items are not sounds, but signs. In the standart model of DM (cf. HALE & MARANTZ, 1993; MARANTZ, 1997) the roots are non-categorial, it means that they do not have a grammatical category and will be formed into noun, verb or adjective when they are c-commanded by a categorizing head. In some models, as the one we are adopting in this paper (cf. MINUSSI, 2008, 2009), the roots are considered to be abstract and must contain semantic features which will form a more general concept. In the model presented in Minussi (2009), as well as in other models of the theory (cf. ARAD, 2003, 2005), there are categorizing heads that, when they c-command abstract roots, they will negotiate a more specific meaning with the roots and they will give then a category. Taking into account the studies developed by Felipe (2006), we are considering that a sign in LSB is formed by features as: handshape (H), location (L), movement (M), orientation or directionality (O/Dir), facial and body expressions (Exp). However, if we take into account DM theory, we need to define what the parameters shown above correspond to in this theory: (i) are they features that belong to the root? or (ii) are they abstract morphemes? The first hypothesis that we suggest in this paper is that these parameters are abstract features that may form an abstract root. However, the abstract roots can be divided into two different kinds: (i) roots in which the feature [+movement] is incorporated to them and (ii) roots in which there is no incorporation of the feature [+movement]: (3)

a)

√+H +L +M +O/Dir +Exp

b)

x ru x [+M]

√+H +L +M +O/Dir

We can find signs9 in which the movement is incorporated into the root and it does not have a direct relation with the category of the sign, in a way that this movement takes part of the concept of such sign. This characteristic can be observed in data as in (4), that will have a structure as in (3)a. (4) 9

a) ÁGUA

WATER

In this paper, we understand “sign” and “word” as the same thing.

4

b) TELEVISÃO c) JANELA

TELEVISION WINDOW

A characteristic of the signs in (4) is that they do not have a verbal counterpart of the sign with neuter lexical category formed with the addition of movement. It means that the verb “AGUAR” (TO-WATER) in LSB is not formed with more movement, but it is formed by adding other element, as other sign or classifier10, for example, as in (5): (5)

a) ÁGUA^REGADORCL (aguar) b) TELEVISÃO ASSISTIR c) ABRIR-JANELACL

WATER^WATERING-CANCL (to water) TELEVISION TO-WATCH (to watch TV) OPEN-WINDOWCL (to open the window)

Although the feature [+movement] do not bring any difference in some words, in other sign languages, as ASL, for example, this feature seems to be essential to distinguish signs that are interpreted as a verb or as a noun, as in (7): (6)

CHAIR / TO-SIT

Supalla e Newport (1978)

A research about the feature [+movement] has already been developed in LSB. Authors as Felipe (2006) show that the difference seen in (7) happens only with some pairs, but not with all of them. In LSB, they are interpreted either as nouns or as verbs and the categorial ambiguity is solved only in the discourse. We have made some tests with LSB signers where we showed a picture and asked them to form a sentence, a context in LSB. In these recordings, we could see as a preliminary result that in fact there is a difference involving movement between the pairs of nouns and verbs for some of the deaf. (7)

a) CADEIRA / SENTAR b) AVIÃO / IR-DE-AVIÃO c) TESOURA / CORTAR-COM-TESOURA A-PAIR-OF-SCISSORS d) CASA / MORAR e) TELEFONE / TELEFONAR

CHAIR / TO-SIT AIRPLANE / TO-GO-BY-AIRPLANE PAIR-OF-SCISSORS / TO-CUT-WITHHOUSE / TO-LIVE TELEPHONE / TO-CALL

The recordings show that for some of the deaf verb formations use to have more movement than noun formations. For example, chair has one movement while to sit has two11 – the same thing is observed in the example in (8)d and in (8)e only the verb has a movement directed to the person I am talking about (it seems to be a kind of agreement); airplane may have short movement while to go by airplane has a longer movement – again the same thing happens with the pair in (8)c. These observations lead us to think about a preliminary hypothesis that the feature [+movement] is the realization of a verbal categorizer when it is not incorporated into the root, making a non-categorial root to turn into a verb.

10

In the sign languages sense as it is explained in footnote number 8.

11

The opposite pattern was observed in Quadros & Karnopp (2004), that is, chair with more movement and to sit with only one movement. We need to apply more tests to learn how these patterns are applied by the deaf.

5

(8)

v ru v [+M]

√+PA [+...]

Then, if we take a pair of signs into account, in which there is a common concept between them as, for example, HOUSE and TO-LIVE, we consider that HOUSE is a noncategorial root, as the feature of movement is not incorporated in it. We also consider that in TO-LIVE, we have the same root HOUSE, that is categorized by a verbal head, realized by the feature of movement. (9)

a)

b)

√ +H +L +O/Dir +Exp HOUSE v = TO-LIVE ru v √ +H [+M] +L +O/Dir +Exp HOUSE

The categorizing head in (10)b, is not only responsible for categorizing the root that has the concept HOUSE, but it also negotiates a new meaning with this root in the sense discussed in the theory of phase in the word, explained in Marantz (2001) and Arad (2003). For the authors mentioned above, phase is not restricted to a vP or CP, but to all and any categorizing head (n, v, a, etc.) that is attached to a syntactic structure. Marantz (2001), intending to reconstruct both places 12 of word formation in syntax, makes a distinction between words that are formed from roots (atomic elements with no functional material) and words formed from existent words, that is, words that are formed from categorized roots. See the scheme in (10): (10)

a)

n, a, v ei n, a, v √

LF PF

12

These two places mentioned may refer to both components of grammar (lexicon vs. syntax) or to two different levels of the same component (derivational level vs. inflectional level). All of the theories share the intuition that there is a lower level in which the word formation is associated to idiosyncrasy (gaps, non-compositional meaning), and other higher level in which word formation would be related to productivity and regularity (that would be the syntactic level, inflectional, etc.) (ARAD, 2003).

6

b)

y ei y n, a, v ei n, a, v √

LF PF

According to Arad (2003), the root is not a real word. It becomes a noun, an adjective or a verb when merged with a head that contains the feature n, a or v, as in (10)a. It would be the first domain or phase if we use the terminology of Chomsky (2001). It is in this domain that the specific meaning of the root is negotiated, accessing the knowledge stored in the Encyclopedia (List 3). In LSB, the categorization phenomenon, at least in a preliminary way, does not seem to occur to all categories. For example, it is not possible to affirm that HOUSE is a noun, as there is no mark that could be considered as a nominal categorizer, with the exception of the lack of movement in relation to its counterpart with movement that is interpreted as a verb, TO-LIVE. We say the same thing about those roots, as WATER, that have a movement feature incorporated to them. It is not possible to say that it is a noun or a verb, as there is no counterpart with or without movement, or with other movement, that is, the verb to water is not derived from the root WATER, from a movement, as we showed in (6)a, this sign is a compound. In the next section, we are going to show the kinds of compounds available in LSB, besides suggesting that there are two kinds of structure to form compounds in this language. 3. Compounding types Felipe (2006) suggests that there are three kinds of compounding in LSB. For this author, the process of composition uses lexical items that are free morphemes that are juxtaposed or agglutinated to form a new lexical item. In this way, the process of compounding can happen through: (i)

(ii)

(iii)

Juxtaposition of two lexical items, that is, two signs that forms a third free form, as: MULHER^BEIJO-NA-MÃO (WOMAN^KISS-ON-THE-HAND) = ‘mãe’ (mother); ASSINAR^SEPARAR (TO-SIGN^SEPARATE) = ‘divórcio’ (divorce). Juxtaposition of a classifier with a lexical item. In this process, the classifier is not a gender mark13 and works as a clitic. Examples of this process are the signs: coisa-pequena^PERFURAR (small-thing^PERFORATE) = ‘alfinete’ (pin); coisapequena^APLICAR-NO-BRAÇO (small-thing^APPLY-IN-ARM) ‘agulha’ (needle); DORMIR^pessoa+14 (SLEEP^people+) ‘alojamento’ (lodging). Juxtaposition dactylology of the word in Brazilian Portuguese with the sign representing the action performed by the noun as in: COSTURAR-COM-

13

The author uses the term gender to refer to things as shape, not to refer to male or female. The symbol + is used by the author in order to represent the plural mark that may happens through the repetition of the lexical item in LSB. 14

7

AGULHA^A-G-U-L-H-A ‘agulha’ (needle)15.

(SEWING-WITH-NEEDLE^N-E-E-D-L-E)

In Figueiredo-Silva & Sell (2010), the authors show three kinds of compounds. The first kind is called “apparent compounds” as the ones in (11): (11)

a) b)

HOMEM^BEBÊ / MULHER^BEBÊ [bebê – menino ou menina] MAN^BABY / WOMAN^BABY [baby – boy or girl]16 HOMEM^TI@ / MULHER^TI@ [tio / tia] MAN^UNCLE / WOMAN^UNCLE [uncle / aunt] (FIGUEIREDO-SILVA & SELL, 2010, p. 17)

The authors consider that nouns in LSB do not have morphological mark for gender. Besides, the combinations in (11) cannot be considered true compounding, because the order of the signs may vary as we observed in (13). According to the hypothesis of compounding that the authors consider, the components order must be invariable. (12)

a) b)

BEBÊ^HOMEM / BEBÊ^MULHER [bebê – menino ou menina] BABY^MAN / BABY^WOMAN [baby – boy or girl] TI@^HOMEM / TI@^MULHER [tio / tia] UNCLE^MAN / UNCLE^WOMAN [uncle / aunt] (FIGUEIREDO-SILVA & SELL, 2010, p. 18)

There are also compounds called “true compounds”, which presents a fixed order. (13)

a) b) c) d)

HOMEM^RURAL [agricultor] MAN^RURAL [agriculturist] MULHER^COSTURA [costureira] WOMAN^SEW [dressmaker] CASA^ESTUDO [escola] HOUSE^STUDY [school] CASA ^CRUZ [igreja] HOUSE^CROSS [church] (FIGUEIREDO-SILVA & SELL, 2010, p. 21 e 22)

Finally, Figueiredo-Silva & Sell (2010) describe a third kind of compound, called “sentence compounds”, which are formed by more than two signs. See the examples below: (14)

a) b)

HOMEM^CONSERTO^ELETRICIDADE [eletricista] MAN^FIX^ELECTRICITY [electrician] HOMEM^ ELETRICIDADE^CONSERTO [eletricista] MAN^ELECTRICITY^FIX [electrician] (FIGUEIREDO-SILVA & SELL, 2010, p. 23)

The authors affirm that not all of the signs formed by two or more signs together are a compound. In the examples in (15), the order is not fixed and there is no obligation of the signs. For example, the sign that refers to ‘electrician’ can be formed without information about gender, that is FIX^ELECTRICITY. Then, these signs are called “apparent sentence compounds”.

15

This is our translation into English.

16

The glosses were tranlated by us.

8

Although, the authors call attention that there is a group formed by two or more signs that uses the form [HOUSE + N + N] to express places/locations and that these can be called compounds, as they have a fixed order and all signs must be realized. (15)

a) b)

CASA^VENDA^PAPEL HOUSE^SELL^PAPER CASA^VENDA^CARNE HOUSE^SELL^MEAT

[papelaria] [stationery] [açougue] [butchery] (FIGUEIREDO-SILVA & SELL, 2010, p. 24)

About this last kind of compound, the data we got for this paper show that it neither seem to have such a regularity about the form HOUSE + N + N, nor about the order of elements that the authors suggest. The sign ‘butchery’, for example, is sometimes realized with the sign STORECL instead of HOUSE. Besides, in our recordings we could observe the sign “butchery” only as STORE^MEAT. These facts suggest that other tests must be done with a greater amount of deaf people, as it may have a dialectal variant affecting this group of compounds. Besides, they reveal that the obligatoriety and order of elements criteria should not be used as as the main points to define the composition phenomenon in LSB. Then, in this paper, we treat all the examples shown above as compounds. In this way, we start discussing a first characteristic concerning compounding using the notion of root discussed above. We may notice that the compounds, in general, have two structures: [√ROOT (neuter) + classifier] and [√ROOT (neuter) + √ROOT (neuter)]. This first hypothesis about the kind of structure of compounds in LSB comes from the observation that there is no compound formed by verbs, and here we understand that a verb is a √ROOT (neuter) + Movement, that is, a √ROOT (neuter) that was categorized, and such categorizer is expressed through movement. Even a root that has an incorporated movement in it, as WATER, may have its movement reduced17 when used to form a compound. In this way, the feature [+movement] seem to be an important feature in compounding phenomenon. The first structure hypothesized, [√ROOT (neuter) + classifier] or [classifier + √ROOT (neuter)], is found in signs as: (16)

a) b) c)

√ÁGUA^[+ CÍRCULOCL] √WATER^[+CIRCLECL] √ÁGUA^[+ QUADRADOCL]18 √WATER^[+SQUARECL] [LOJACL]19^√PIZZA [STORECL]^√PIZZA

[lagoa] [lake] [piscina] [swimming pool] [pizzaria] [pizzeria]

At this point, we call classifier not exactly what is considered a classifier in LSB, but we use the idea of classifiers in oral languages. Some languages, like Miraña, use certain markers concerning the shape of an object, indicating that the object is round, elongated, thin, square, to form new words. We can observe some examples of Miraña in (17): 17

This reduction of movement can be compared to the phenomenon of loss of accent that happens in some words in oral languages, when they take part in a compound formation. 18 We also found this sign as QUADRADOCL^NADAR (SQUARECL^SWIM) and ÁGUA^QUADRADOCL^NADAR (WATER^SQUARECL^SWIM). 19 We found the sign STORE as a classifier (the form of a big door – handshape in “1”) and as a sign – handshape of active hand (the hand that produces the sign, with movement) in “L” moving in circle beside the passive hand, and passive hand (a hand that works as basis, with no movement) in “S”.

9

(17)

a)

b)

Pihhúi-ko fishery-specific class marker (SCM).1D20.pointed ‘fishing rod túia-hui foot-SCM. orifice ‘shoe’

In (17)b), for example, we have a root túia, which means ‘foot’ and a specific which means ‘orifice’. Together, the root and the class marker, they mean ‘shoe’. Thus, we can consider that features like [+square], [+circle] in LSB are equivalent to some class marker in oral languages. Vale ressaltar que uma das características dos classificadores da LIBRAS é não significar algo sozinho, de modo que um classificador sempre está atrelado a um raiz por exemplo, como um afixo ou uma base presa na tradição morfológica. The second structure that we may found in compounds is the most traditional one, in which there are two roots taking part in the compound, [√ROOT (neuter) + √ROOT (neuter)], as in: (18)

a) b) c) d)

LOJA^PIZZA CASA^ESTUDAR CASA^CRUZ MORTE^CRUZ

[pizzaria] [escola] [igreja] [cemitério]

STORE^PIZZA HOUSE^TO-STUDY HOUSE^CROSS DEATH^CROSS

[pizzeria] [school] [church] [cemetery]

These compounds have the following structure: (19)

√CHURCH ei √ +H √ +H +L +L +O/Dir +O/Dir +Exp +Exp HOUSE CROSS

As we have mentioned before, we consider that the compounds in LSB are formed by neuter (non-categorial) roots, instead of verbs. This characteristic is more evident in compounds in which there is a verbal counterpart of a root, as it happens with HOUSE and TO-LIVE. A compound as “church” is not formed by TO-LIVE^CROSS, once the first element lacks the feature [+movement]. 4. Other kinds of compounds Meir et al. (2010) showed that in ASL some compounds may be formed with simultaneous signs. Since LSB has two independent articulators, the hands, and a few signs produced by facial expressions only, as SEXO (SEX) and LADRÃO (THIEF), forming simultaneous compounds is a logical possibility in this language, what doesn’t happen in oral languages. Then, in LSB, as well as in ASL, we can separate compounds in: nonsimultaneous and simultaneous. The non-simultaneous compounds are the ones we have presented so far, as: 20

1D (one dimension).

10

(20)

a. CASA^ESTUDAR (escola) b. ÁGUA^CAMINHO (rio) c. LOJACL^PIZZA (pizzaria)

HOUSE^STUDY (school) WATER^PATH (river) STORECL^PIZZA (pizzeria)

The simultaneous compounds presents two different signs realized at the same time, as in: (21)

a. SEXO^VIAJAR (lua-de-mel) SEX^TRAVEL (honeymoon) b. LADRÃO^ATRAIR (“ganhar o público”) THIEF^ATTRACT (“to call attention of the public”) c. PÉ^PÉCL (chinelo) FOOT^FOOTCL (flip flop)

We can also observe that examples (21)a/b bring signs formed by one sign that is realized by facial expression (SEX in (21)a and THIEF in (21)b) and one sign realized by the hands (TRAVEL in (21)a and ATRACT in (21)b). So, these simultaneous compounds are formed by one sign realized by facial expression and other one realized by one hand. The example in (21)c, on the other hand, shows a compound realized by each hand forming a different sign. This is a possibility when movement is not involved, since a human being wouldn’t be able to make a different sign with a different movement using each hand simultaneously, for motor reasons21. The class of non-simultaneous compounds, as we can see, have one non-compositional meaning, namely, it creates a new meaning that is more than the sum of the meanings of the component signs. Another attribute that makes this class one particular class of compounds is the presence of movement into the roots forming compounds. This fact must not be linked to the compositional nature of the compound, so that the movement should not be taken into account for the formation of a new meaning of the compound. 5. Conclusion As results for this research, we examined the kinds of productive compounds in LSB and identified some properties present in these compounds. The hypotheses raised here are the following: i)

ii)

By using the theoretical framework of DM, we argued that there are two types of roots: (i) roots in which the movement feature is incorporated and (ii) roots in which the movement feature is not incorporated. Compounds in LSB may be formed by two signs. These signs are formed by non-categorial roots, that is, roots that have a movement feature incorporated into them, and one classifier; or by roots in which there is no movement feature incorporated and one classifier. The later type of roots can form verbs by adding the movement feature to them.

We also showed that there is a special kind of compounds in LSB which have two movements, one made by the hand and the other one made by the tongue (a sign formed by 21

An anonymous reader called our attention to the possibility of considering ESCREVER as a compound, once it is formed by a passive hand forming a paper shape using a Classifier (in sign languages sense) and an active hand making the movement of “writing”. There are other examples of this kind as LER (READ), DESENHAR (DRAW) among others, but this discussion may be developed in future research.

11

only a facial expression, as we saw above). These are called simultaneous compounds and in these examples the movement should not be taken into account for the formation of a new meaning of the compound. 6. References BRITO, Lucinda Ferreira. Por uma gramática de línguas de sinais. 2. ed. Rio de Janeiro: Tempo brasileiro, 2010. [1995]. CAPOVILLA, Fernando César; RAPHAEL, Walkiria Duarte. Dicionário enciclopédico ilustrado trilíngue da Língua de Sinais Brasileira – Libras. v. I e II. São Paulo: Editora da Universidade de São Paulo/ Imprensa Oficial do Estado, 2001. DEL GIUDICE, Alex. A constraint-based analysis of compound formation in the sign modality. 2007. Disponível em: http://idiom.ucsd.edu/~delgiudice/DelGiudiceASLcompounds.pdf. Acesso em: Acesso em: 10 nov. 2011. FELIPE, Tanya Amara. Os processos de formação de palavras na LIBRAS. ETD – Educação Temática Digital, Campinas, v. 7, n. 2, p. 200-217, jun. 2006. ______. A relação sintático-semântica dos verbos na Língua Brasileira de Sinais (LIBRAS). 1998. Tese (Doutorado) – Universidade Federal do Rio de Janeiro, Rio de Janeiro. v. I e II. FIGUEIREDO SILVA, Maria Cristina; SELL, Fabíola Ferreira Sucupira. Algumas notas sobre os compostos em português brasileiro e em LIBRAS. PPT apresentado na USP e artigo disponibilizado por e-mail, 2009. KLIMA, Edward S.; BELLUGI, Ursula. The Signs of Language. Cambridge, MA: Harvard University Press, 1979. LEITE, Tarcisio de Arantes. A segmentação da língua de sinais brasileira (libras): um estudo lingüístico descritivo a partir da conversação espontânea entre surdos. 2008. Tese (Doutorado) – Universidade de São Paulo, São Paulo. LIDELL, Scott K. THINK and BELIEVE: sequentiality in American Sign Language. Language, v. 60, n. 2, p. 372-399, 1984. ______; JOHNSON, Robert E. American Sign Language compound formation processes, lexicalization, and phonological remnants. Natural Language and Linguistic Theory, v. 4, n. 8, p. 445-513, 1986. MCCLEARY, Leland Emerson; VIOTTI, Evani de Carvalho. Transcrição de dados de uma língua sinalizada: um estudo piloto da transcrição de narrativas na língua de sinais brasileira (LSB). In: SALLES, H. (Org.). Bilinguismo e surdez. Questões linguísticas e educacionais. Goiânia: Cânone Editorial, 2007. P. 73-96. MEIR, Irit; ARONOFF, Mark; SANDLER, Wendy & PADDEN, Carol. 2010. Sign language and compounding. In SCALISE, S. & VOGEL, I. (Eds.). Compounding. John Benjamins, p. 301-322. MINUSSI, Rafael Dias. Os nomes compostos no português brasileiro: uma análise morfossintática. Colóquio de Morfologia. 2011. Disponível em: http://coloquiodemorfologia.files.wordpress.com/2011/05/minussi.pdf. Acesso em: 10 nov. 2011. ______. Os nomes compostos do hebraico: uma análise morfossintática. Estudos Linguísticos, São Paulo, v. 38, n. 1, p. 259-271, jan./abr. 2009. QUADROS, Ronice Muller; KARNOPP, Lodenir Becker. Língua de sinais brasileira: estudos lingüísticos. Porto Alegre: Artmed, 2004. SIDDIQI, Daniel. Syntax within the word: economy, allomorphy, and argument selection in Distributed Morphology. Linguistik Aktuell/Linguistics Today 138. Amsterdan: John Benjamins, 2009 12

STOKOE, William. Sign language structure: An outline of the visual communication systems of the American Deaf. Studies in Linguistics, Occasional Papers 8. Buffalo: University of Buffalo Press, 1960. SUPALLA, Ted; NEWPORT, Elissa. How many seats in a chair? The derivation of nouns and verbs in American sign language. In: SIPLE, P. (Ed.). Understanding language through sign language research. New York, Academic Press, 1978. UYECHI, Linda. Merging in American Sign Language compounds. In: DUNCAN, E.; FARKAS, D.; SPAELTI, P. The Proceedings of the Twelfth West Coast Conference on Formal Linguistics. 1993. p. 167-182. XAVIER, André N. Descrição fonético-fonológica dos sinais da língua de sinais brasileira (LIBRAS). 2006. 175f. Dissertação (Mestrado) – Universidade de São Paulo, São Paulo.

13