Chapter Irregular Negatives in Sign Languages

by Ulrike Zeshan

Please note: A common introduction to Chapters 139 and 140 on Sign Languages is available.

1. Defining the values

The map shows the distribution of irregular negatives in sign languages with respect to the number of such forms in each language. The following values are represented in the map:

Values of Map 139A. Irregular Negatives in Sign Languages
Go to map
No irregular negatives 1
One irregular negative 3
Some irregular negatives (2-5) 10
Many irregular negatives (more than 5) 21
Total: 35

The term irregular negative refers to several types of forms that will be called negative suppletion and negative derivation here. They all have in common that they express negation in a non-standard way (see Zeshan 2004b for details).

The canonical marking of negation in sign languages involves, on the one hand, negative particles that are independent signs and, on the other hand, non-manual negative signals (head movements and facial expressions) that co-occur with the negated signs. Neither the negative particles nor the non-manual negative signals affect the form of the sign or signs being negated. In example (1), from Indo-Pakistani Sign Language, there is a clause-final negative particle, and the non-manual negative signal is a headshake (indicated by the line labelled “neg” on top of the sign glosses). The negated predicate ‘interesting’ takes the same form as it would in a positive clause.

(1) Indo-Pakistani Sign Language (own fieldwork) 







‘This book is not interesting.‘

In the case of irregular negatives, on the other hand, the negation is part of the negated sign in one way or another, so that the negative form looks different from the positive form. These negative forms are called irregular because their number is always very small in any given sign language, with a maximum of around 25 items. Even if some of the negatives are consistently derived from positive forms in the same way, as happens in negative derivation, this is still a very limited process and quite different from the situation found in many spoken languages where a negative affix can be added to all verbs or predicates.

Negative clauses using irregular negatives usually cannot include any additional independent negative particle, but are otherwise no different from regular clause negation. Nonmanual negative signals combine with irregular negatives in the same way as with canonical negation (example 2, from South Korean Sign Language).

(2) South Korean Sign Language (K.-S. Byun, p.c.)





‘(I) don’t know the way.’ 

Negative suppletion involves negative forms that are entirely different from the corresponding positive forms. For example, the signs WANT and WANT-NOT in Russian Sign Language look entirely different and cannot be related to each other by any formational process (see Figure 1). Negative derivation, on the other hand, involves a set of negative forms that are derived from their positive counterparts in a regular way. There are two different types of negative derivational processes in sign languages. Negative derivation may take the form of internal modification of a sign, typically in the form of a modified movement pattern that is superimposed on a sign in a simultaneous way. For example, a set of eight negative signs is derived in DGS (German Sign Language) by adding a downward and diagonal inward-outward movement pattern to the positive sign (see Figure 2). The second type of negative derivation is sequential rather than simultaneous and is more comparable to negative affixing in spoken languages. For example, in a number of sign languages an outward twist of the wrist together with an opening of the hand is added to a positive sign to form a negative. The negative sign always ends in a position with the palm facing forward and all fingers extended. Figure 3 shows an example from Ugandan Sign Language. Another type of irregular negative, involving a negative handshape, is much less common across sign languages than negative movement patterns. A handshape with only the little finger extended, meaning BAD when used by itself, occurs in some East Asian sign languages and is used for negative derivation of both the simultaneous and the sequential type. The latter is exemplified in Figure 4, showing the sign KNOW-BAD ‘not know’ in Hong Kong Sign Language.

Figure 1: WANT
(Russian Sign Language)


Figure 2: CAN
(German Sign Language)


Figure 3: LIKE
(Ugandan Sign Language)


Figure 4: KNOW-BAD
(Hong Kong Sign Language)

The typological distinction between a sign language with “some irregular negatives” and a sign language with “many irregular negatives” is not only a matter of the number of items, but correlates with another typological parameter. In the former type, all of the irregular negatives are suppletive forms, while the latter type involves both negative suppletion and negative derivation. This means that some of the negatives are formationally related by taking part in the same derivational process and thus forming a paradigm of derived negatives. The presence or absence of negative derivation, and the associated paradigmatic organization of related negatives, thus correlates with the number of irregular negatives in the following way:

Some irregular negatives (2-5): 

  • negative suppletion only  

  • no paradigmatic organization 

Many irregular negatives (>5): 

  • negative suppletion and derivation 

  • paradigm (s) of related derived negatives

On the other hand, it does not matter for the purposes of this typology whether a language makes use of one or several derivational paradigms, how large the paradigm is, and which formational processes are applied to derive the negative forms. Finally, the category with “some irregular negatives” also includes a few borderline cases where no clear evidence has been found for the existence of a larger number of irregular negatives and/or the existence of a paradigm of forms related by a common negative derivational process. 

2. Geographical distribution

Few generalizations emerge from the map in terms of areal patterns. Worldwide, irregular negatives are a common phenomenon across sign languages, with the majority of languages in the category having many irregular negatives. However, there is a striking absence of irregular negatives in the South Asian subcontinent, with a Pakistani variety entirely lacking such forms, and only a single irregular negative in Indian varieties of Indo-Pakistani Sign Language. Another sign language with just one irregular negative is Lengua de Señas Española (Spanish Sign Language). It is notable that the single irregular negative is a negative existential in both these cases. Moreover, the village-based sign languages are also strikingly poor in irregular negatives, with a single suppletive negative (aspectual NOT-YET) in Kata Kolok and only two irregular negatives in Adamorobe Sign Language.

For the most part, rather than being an areal phenomenon, the extent of irregular negativization seems at least partly to be linked to language families, although the notion of language family and genealogical grouping is itself unclear for sign languages (see introduction to the sign language chapters). Both the British Sign Language (BSL) family and the French Sign Language Langue des Signes Française (LSF) family have many irregular negatives. The BSL family also includes Auslan (Australian Sign Language) and New Zealand Sign Language. The LSF family also includes American Sign Language, Langue des Signes Québecoise, and Russian Sign Language. This group of sign languages also accounts for six of the 11 sign languages with a particularly high number of irregular negatives, amounting to 10 items or more.

European sign languages as a whole predominantly have many irregular negatives with some negative derivation, and this is also true of their derivatives in other parts of the world, including North and South America, Australasia, Israel, and, indirectly via American Sign Language, Uganda. A particularly widespread type of negative derivation, involving an outward twist of the wrist and an open hand (see Figure 3), is also limited to European and European-based sign languages, occurring in Australia, New Zealand, France, Greece, Finland, Russia, Uganda, Brazil, Québec, the United States, and Israel. For Icelandic Sign Language, Irish Sign Language and Lingua Italiana dei Segni, it has not been possible to establish the existence of a larger number of irregular negatives or a negative derivational process. The latter two sign languages do have individual forms that appear as part of larger paradigms in other sign languages, but in the absence of any other forms following the same patterns, they do not qualify as negative derivation in these sign languages.

There are a number of other such cases where signs that are part of a larger paradigm in other sign languages appear as individual irregular negative signs, and thus seem to have been borrowed. Borrowing and language contact do play a role in sign languages, but the contact situations that arise are much more complicated than is typically the case with spoken languages (see introduction to the sign language chapters). It does seem that while individual irregular negative forms are readily borrowed almost on a worldwide level, derivational processes are not generally borrowed as a whole. This may account for a number of cases where we have a larger set of derivational irregular negatives in one sign language and one or two identical forms that are not part of a similarly large paradigm in another sign language. For example, Russian Sign Language has two signs IMPOSSIBLE and NOT-NECESSARY involving the same negative movement pattern as in German Sign Language (DGS), but lacks a more extensive paradigm (eight signs in DGS). American Sign Language has a larger paradigm of five signs with the same outward twist of the wrist, while Ugandan Sign Language has a paradigm with fewer members, including KNOW-NOT and LIKE-NOT, but not HAVE-NOT.

Sign languages in East Asia also have a large number of irregular negatives, but the most common derivational processes are different from those in European sign languages and their derivatives. There is extensive use of a negative handshape (see Figure 4) in Chinese Sign Language and, to a lesser extent, in Hong Kong Sign Language, which may be a related sign language. Reversal of movement for negativization is used in both these sign languages and in Nihon Shuwa (Japanese Sign Language). East Asian sign languages form language families of their own, but are not related to sign languages outside the region.

In addition to language families and language contact, there are many other factors that influence the number of irregular negatives. It seems that older sign languages are more likely to have many irregular negatives than sign languages that are very young or are the result of recent creolization of an indigenous and a foreign sign language. LSF, which is rich in irregular negatives, is one of the oldest documented sign languages. On the other hand, Thai Sign Language, which has only some suppletive irregular negatives, may have had a recent history of creolization due to influence from American Sign Language. Türk İşaret Dili (TID), however, seems to have had a longer history of its own, with evidence of advanced grammaticalization especially in the domain of negatives. This involves extensive negative cliticization, but only a few suppletive negative forms; many items that appear as irregular negatives in other sign languages participate in negative cliticization in TID instead, thus reducing the number of irregular negatives. While the age and historical development of Lughat al-Isharat al-Lubnaniya (Lebanese Sign Language) is unknown, Tanzania Sign Language is definitely one of the young sign languages in the sample. This sign language does have several irregular negatives, partly due to the influence of gesture. If the local culture uses any negative gestures, these may be grammaticalized and incorporated into the sign language as signs, and this has happened with two negative gestures in Tanzania Sign Language. However, this sign language lacks any negative derivational processes.

Irregular negatives typically occur in a number of specific semantic and grammatical domains. These include the following (with examples):


not know, not understand

emotional attitude:

not want, not like, not care


cannot, need not, must not


not have, not exist, not get


will not, did not, not finished

evaluative judgement:

not right, not possible, not enough

For sign languages that have only some irregular negatives (5 or less), these are consistently drawn from a very restricted sub set of this list, which are at the same time the most frequent irregular negatives across all sign languages. These include ‘want’, ‘like’, ‘know’, ‘can’, ‘must’, ‘finish’ (completive) and ‘exist/have’ (existential). The last and vaguest category of “evaluative judgement” is the only domain that is not represented among the most frequent items, and is at the same time the domain where there is most variation across sign languages.