Nat Lang Linguist Theory (2018) 36:685–699
DOI 10.1007/s11049-017-9389-5
Visible degrees in Italian Sign Language
Valentina Aristodemo1 · Carlo Geraci1
Received: 15 March 2016 / Accepted: 25 July 2017 / Published online: 14 November 2017
© Springer Science+Business Media B.V. 2017
Abstract Data from Italian Sign Language provide evidence in favor of a degreebased
analysis over a non-degree based analysis for gradable adjectives and comparative
constructions. Morphological and phonological constraints identify a class of
gradable adjectives in which degree variables can be overtly represented as ordered
points established in the signing space by an iconic mapping. When this happens
the visible degree becomes available as an antecedent for a later pronoun, as in the
nominal, temporal and modal domain, showing that the same anaphoric system is at
work.
Keywords Degrees · Gradable adjectives · Comparatives · Sign Language ·
Iconicity · Degree anaphora
The research leading to these results received funding from the European Research Council under the
European Union’s Seventh Framework Programme (FP/2007-2013) / ERC Grant Agreement
N°324115 FRONTSEM (PI: Schlenker). Research was conducted at Institut d’Études Cognitives,
École Normale Supérieure – PSL Research University. Institut d’Études Cognitives is supported by
grants ANR-10-LABX-0087 IEC and ANR-10-IDEX-0001-02 PSL. This research has been possible
also thanks to the SIGN-HUB project, which has received funding from the European Union’s
Horizon 2020 research and innovation programme under grant agreement N. 693349 (PI: Carlo
Geraci) and to the Fyssen Foundation Research Grant Historical relations across sign languages:
Language Families or Iconic similarities? (PI: Carlo Geraci)
B V. Aristodemo
vale.aristodemo@gmail.com
C. Geraci
carlo.geraci76@gmail.com
1 Institut Jean-Nicod, Département d’Études Cognitives, CNRS, ENS, EHESS, PSL Research
University, 75005 Paris, France
686 V. Aristodemo, C. Geraci
1 Introduction
Sign languages systematically make visible some abstract properties of human language
that are not immediately visible in spoken languages (Wilbur 2008). For example,
loci, which are positions in signing space associated with nominal elements, have
been claimed to be overt realization of indices, i.e. of variables (Sandler and LilloMartin
2006). Schlenker (2013) uses sign language data to prove the existence of
time and world variables. He shows that in American Sign Language (ASL) temporal
and modal constructions can be anaphorically retrieved by a pronoun, realized as a
pointing sign toward a locus thereby supporting the existence of time and world variables.
The example in (1) taken from Schlenker (2013:6) shows that the two spatial
locations associated with the temporal constructions TOMORROW and DAY-AFTERTOMORROW
can be singled out by the pointing pronoun IX.
(1) a[TOMORROW WILL RAIN] b[DAY-AFTER-TOMORROW WILL SNOW].
re re
IXb IX-1 HAPPY. IXa IX-1 WON’T HAPPY.
‘Tomorrow it will rain and the day after tomorrow it will snow. Then [= the
day after tomorrow] I will be happy but then [= tomorrow] I won’t be happy.’
Our work fits into the context of these studies. We look at gradable adjectives and
comparative constructions in Italian Sign Language (LIS). Specifically, we argue that
locations in space associated with gradable adjectives are loci and correspond to overt
realizations of degree variables. Our proposal is in part reminiscent of Wilbur et al.
(2012), where intensification in gradable adjectives in ASL is discussed. In that paper
the authors propose that phonological boundaries of gradable adjectives iconically
represent the boundaries of degree scales. By looking into LIS, we will show that the
scale itself is visible and not just its extremes and that LIS phonological boundaries
mark the standard and the reference degree.
Formal approaches to the semantics of gradable adjectives are distinguished between
those that make explicit reference to degrees in the ontology, and those that do
not. The data we present in this short paper support a degree-based approach. We describe
how gradable adjectives and comparatives are expressed in LIS, showing that
degrees can be overtly realized as points in the signing space (i.e. loci). When this
happens, anaphoric relations between the visible degree and a pronoun become accessible.
Similarly to temporal and modal anaphora in ASL, anaphoric constructions
can involve singular, dual or trial pronouns, showing that in these domains the same
pronominal system is at work (Schlenker 2013).
The rest of the paper is organized as follows: Sect. 2 briefly introduces the two
main accounts of gradability. In Sect. 3, we present the empirical part of our work
describing new data on gradable adjectives and comparative constructions in LIS. We
also show how the pronominal system interacts with gradable adjectives. Section 4
contains the analysis, while Sect. 5 concludes the paper.
2 Two theories of gradability
There are two main approaches to the semantics of gradable adjectives and gradability:
the degree-based approach or Scalar Analysis (Cresswell 1976; Heim
Visible degrees in Italian Sign Language 687
1985; Kennedy 1997, 2007:etc.), and Delineation Semantics (Klein 1980; Burnett
2015:etc.). For a general overview on the semantics of adjectives and comparatives
see Demonte (2011) and Beck (2011), respectively. The Scalar Analysis treats gradable
adjectives as relations between individuals and degrees, which are abstract representations
of measurements. To do so, degrees and scales are introduced into the
ontology and adjectives are treated as expressions that relate individuals to degrees
on a scale. For instance, the denotation of the adjective tall is stated as in (2), where
the measure function height maps an individual to its degree of height:1
(2) ttallu = λd.λx.height(x) = d
As for the semantics of comparative constructions, the degree morpheme -er is
analyzed as a quantifier over the degree variable introduced by the gradable adjective.
It denotes a relation between two sets of degrees and compares the maximal degree of
each of them (Heim 2001; von Stechow 1984:among others). The formal denotation
is given in (3):
(3) t -er u = λD1<d,t>λD2<d,t>.max(D2) max(D1)
The logical representation of the comparative sentence in (4) is stated in (5).
(4) John is taller than Mary.
(5) max(λd.height(j) d) max(λd’.height(m) d’)
In Delineation Semantics, degrees and scales are not part of the ontology. The interpretation
of gradable adjectives is derived by looking at comparison classes which
are sets of individual, adjective pairs pertinent to the context. For instance, in order
to give the semantics of explicit comparatives like the English example in (4), Burnett
(2015:240) adds the expression ER to the logical language, which compares two
individual, adjective pairs with respect to a strict order definable from the comparison
classes. The sentence in (4), then, is true if there is some way of dividing the set of
individuals into two sets like ‘tall’ and ‘not tall,’ obeying certain logical constraints2
such that John is in the set ‘tall’ and Mary is in the set ‘not tall.’ Notably, under this
analysis, gradability is captured indirectly. The logical form of the sentence in (4) is
given in (6):
(6) ER+ (John, tall, Mary, tall)
Both approaches correctly derive the truth conditions of gradable adjectives and
comparatives. However, they differ in how these are derived. Specifically, the existence
of degrees in the ontology opens the possibility to anaphorically refer back to
them, an option that is not available in Delineation Semantics. To put it in terms of
Quine (1948:7): “[...] to be is to be in the range of reference of a pronoun. Pronouns
1This analysis is identical to that of Cresswell (1976), but this is not the only option. Other analyses assume
a partial ordering relation (Heim 1985).
2These constraints include the following: (i) for a given comparison class X, an individual cannot be both
tall and not tall in X; (ii) for two comparison classes X1 and X2, it can’t be the case that John is tall and
Mary is not tall in X1, but Mary is tall and John is not tall in X2 (Burnett 2015).
688 V. Aristodemo, C. Geraci
are the basic media of reference [...].” In this respect, at a first glance, evidence from
English and other spoken languages seems to point toward Delineation Semantics. Indeed,
in sentences like (7) and (8) the pronoun it cannot refer to the degree of John’s
tallness.
(7) # John is taller than Bill. It is 1.70 meters.
Intended meaning: ‘John is taller than Bill and his height is 1.70 meters.’
(8) # John is tall. It is 1.70 meters.
Intended meaning: ‘John is tall and his height is 1.70 meters.’
However, the reason why pronouns cannot anaphorically refer to degrees in English
may be due to factors other than the fact that degrees are not part of the ontology.
One possibility could be that degrees are part of the ontology, but when they remain
covert they are not a suitable antecedent for pronouns (we come back to this issue at
the end of Sect. 3.3).3 If this is the case, then there should be languages in which overt
pronouns can refer back to degrees. Here, we argue that this is indeed the case of LIS,
where degree variables are visible under iconic uses of signing space, as we will show
in Sect. 3. Our line of argument follows Schlenker (2013), who shows that ASL pointing
pronouns can single out loci to refer back to temporal and modal constructions,
hence supporting the existence of time and world variables (see example (1), above).
We provide empirical evidence supporting a scalar analysis of gradable adjectives by
looking at the anaphoric dependency between a visible/overt degree introduced by
positive and comparative markers and a pronoun referring to it.
3 LIS data
The basic facts about gradable adjectives and comparative constructions in LIS are
provided in this section. We also offer empirical evidence supporting a scalar analysis
of gradable adjectives by looking at the anaphoric dependency between a locus introduced
by positive and comparative markers and a pronoun referring to it. Data comes
from two LIS signers (Mirko Santoro and Lorenzo Laudo) who regularly collaborate
with our group. During elicitation we used LIS as the exclusive means of communication.
All LIS productions by the informants were recorded and played back to
collect acceptability and felicity judgments.
To help the reader, we provide here a small guide to assess the LIS examples.
Glosses for signs are reported in small capital letters, Roman subscripts indicate loci
that correspond to locations of nominal referents, Greek subscripts indicate loci that
correspond to locations of degrees. To illustrate, consider the example in (9), where
the nominal element PIETRO is associated with a single locus a, while the adjective
TALL is associated with two loci α and β which correspond to the standard and reference
degrees. Pointing pronouns are glossed as IX. Therefore, the subscript in IXa
3Another possibility could be that the English pronominal system lacks a pronominal form that refers to
degrees altogether. This situation would be similar to the one described for parasitic gaps targeting prepositional
phrases in English (Engdahl 2001, but see Levine et al. 2001 for counter arguments). Specifically,
Engdahl (2001) argues that differently from Swedish, English does not have parasitic gaps on prepositional
phrases because English lacks proforms that are able to refer to prepositional phrases.
Visible degrees in Italian Sign Language 689
indicates a pronoun that points toward locus a. Following the standard practice in
semantics, we use the gloss pos for the positive degree morpheme. We extend this
practice to neg which indicates the negative degree morpheme and iconic-more and
iconic-less which indicate the iconic version of comparative degree morphemes instantiated
in LIS.
(9) Word by word images of a LIS sentence
PIETROa IXa TALLα-posβ
‘Pietro is tall.’
3.1 Gradable adjectives in LIS: The positive form
The adjectives discussed in this paper are all open scale gradable adjectives. They
are gradable because they are compatible with the degree adverb very as in (10a)
and they are open scale adjectives because they are not compatible with adverbs like
completely as in (10b) (see Kennedy and McNally 2005 for an illustration of these
diagnostics):
(10) a. GIANNI TALL-αposβ VERY
‘Gianni is very tall.’
b. *GIANNI TALL-αposβ COMPLETELY
We classify LIS gradable adjectives in two groups: adjectives in which the amount of
the property is iconically encoded, and adjectives in which the amount of the property
is not iconically encoded. Adjectives that belong to the first class, like those shown in
(11)4 need to meet two crucial requirements: i) they are all classifier signs of the Size
and Shape type, ii) the movement is always perpendicular to the plane of articulation.
(11) a. TALL-αposβ b. DEEP-CL-αposβ
4The sign for deep is glossed DEEP-CL rather than DEEP because in the lexicon of LIS there are two signs
for deep. One is a classifier sign while the other is not (see also the discussion around the example (14a)
in the text).
690 V. Aristodemo, C. Geraci
c. CULTURED-αposβ d. BIG-αposβ
First, a morphological constraint is at work. Adjectives of the first group are all classifier
signs, although some of them like those in the examples in (11) and (12) may
have become lexicalized signs, i.e. part of the core lexicon of LIS. More specifically,
they are Size and Shape Specifiers (Supalla 1982). What distinguishes these classifiers
from other types of classifiers, like handling or entity classifiers, is that they
depict the size and/or the shape of the concept they describe by representing the contour
of the object. For instance, the handshape of TALL classifies the area on top of
the subject of predication, the handshape of DEEP-cl is a point-size classifier normally
used for round/cylindrical cavities (like holes, wells, etc.) and the handshape
of BIG classifies the shape of the object of the predication, etc. Differently from Size
and Shape Specifiers, handling classifiers identify objects by the way they are handled,
for example a closed fist handshape is used in many sign languages to represent
hammers, tennis rackets and pans; while in entity classifiers, the handshape itself represents
the entire object/concept, for example a flat open hand is used in LIS to refer
to (non-flying) vehicles (car, trucks, boats, etc.).
Second, a phonological constraint is also active. The movement of the sign has to
be perpendicular with respect to the plane of articulation. The movement component
of signs is normally described as being parallel with respect to the plane of articulation
or as perpendicular to it (Brentari 1998). To illustrate, consider the signs LONG
and CULTURED in (12) and (11c). Both are articulated on some body location, LONG
is articulated on the chest, while CULTURED is articulated on the forehead. However,
in the case of LONG the movement is parallel to the body (i.e. the hand moves
vertically along the chest, which is the plane of articulation), while in the case of
CULTURED the movement (finger aperture) is perpendicular with respect to the plane
that contains the forehead. The same phonological distinction is available for signs
produced in the area in front of the signer. Specifically, the signs in (11) all instantiate
a movement that is vertical with respect to their plane of articulation, as indicated in
the still images. As we will shortly see, the grammar of LIS (and possibly of other
sign languages) capitalizes on this fact to make the scale iconically visible, i.e. only
movements that are perpendicular to the relevant plane of articulation may make the
scale iconically visible.
The second class of adjectives is negatively defined as adjectives that cannot iconically
map the amount of the property onto the signing space. This is so because
they violate either the first (they are not Size and Shape Specifiers) or the second
(the movement is not perpendicular) requirement. The relevant examples are given
in (12)–(14). LONG is a Size and Shape Specifier with no perpendicular movement
(cf. (12)). TALL-entity is not a Size and Shape Specifier (cf. (13)). DEEP and SMART
are non-classifier gradable adjectives with perpendicular movement (cf. (14)).
Visible degrees in Italian Sign Language 691
(12) LONG (13) TALL-entity
(14) a. DEEP b. SMART
Notice that adjectives that are intuitively ‘iconic’ may fail to meet one or both
of these morpho-phonological requirements. For example, DEEP is an ‘iconic’ sign
because in its articulation the dominant hand goes down beyond the non-dominant
hand representing the idea of depth. However, it is not a Size and Shape Specifier
because it does not depict the contour of any object or concept. It is normally used
to indicate the depth in wide spaces. On the other hand DEEP-cl is a Size and Shape
Specifier because it describes the contour of the object, which in the case of cavities,
is the interior. Therefore it can only be used to indicate the depth of cavities, like
holes, wells, the interior of pipes etc. In these latter cases, DEEP cannot be used.
Thus, iconicity alone is not a sufficient condition to encode a visible degree and is a
separate property from iconically encoded scales.
In the next sections, we provide independent evidence for this classification by
showing that only adjectives that belong to the first class can express comparative
constructions by using the synthetic form (Sect. 3.2), and that only for these adjectives,
anaphoric reference to degrees is possible (Sect. 3.3). In Sect. 4, we argue that
when these morpho-phonological properties are met, the amount of the property is
iconically encoded by the movement component which marks two points in the signing
space.
3.2 Gradable adjectives in LIS: The comparative form
The sentences in (15a) and (15b) exemplify the two main strategies to convey more-
comparatives:
(15) a. MAN TALL-αposβ WOMAN MORE
b. MAN TALL-αposβ WOMAN TALL-βiconic-moreγ
‘The woman is taller than the man.’5
5For convenience we translate the meaning of the sentence with an English than-clause even if the syntactic
structure of the LIS sentence is different from the than-clause.
692 V. Aristodemo, C. Geraci
In (15a) the comparison is expressed by an analytic form using the lexical comparative
marker MORE shown in (16), while in (15b) the comparison is expressed by
a synthetic form that we glossed iconic-more. Since this is a simultaneous bound
morpheme, we exemplify it in (17) with the adjective TALL.6
(16) MORE
(17) a. TALL-αposβ b. TALL-βiconic-moreγ
Both classes of adjectives allow comparatives to be expressed with the analytic
form (as in (15a) and (18a)/(19a)). Crucially, only the adjectives in the first class
allow the synthetic form, as shown by the ungrammaticality of (18a) and (19a):
(18) a. ADRIATIC DEEP, AEGEAN MORE
b. *ADRIATIC DEEP AEGEAN DEEP-αiconic-moreβ
‘The Aegean sea is deeper than the Adriatic sea.’
(19) a. MAN SMART, WOMAN MORE
b. *MAN SMART WOMAN SMART-αiconic-moreβ
‘The woman is smarter than the man.’
Less-comparatives behave similarly: they can be expressed by the analytic form
using the lexical sign LESS or by a synthetic form glossed iconic-less as shown in
(20).7
6Iconic-more behaves like the positive morpheme pos, it iconically maps the amount of the property denoted
by the subject (the man and the woman in the case of (15b)). Specifically, the iconic mapping of the
comparative morpheme iconic-more needs to be the same as that of the positive morpheme pos (a vertical
line in the case of TALL, a horizontal in the case of BIG, etc.)
7The main difference between iconic-more and iconic-less is that the movement of the latter goes in the
opposite direction. More generally, in more-comparatives the movement of iconic-more goes in the same
direction as that of the positive form. Hence, if the movement of the positive form is directed downward,
as in DEEP-cl, then iconic-more is also directed downward.
Visible degrees in Italian Sign Language 693
(20) a. MAN TALL-αposγ WOMAN LESS
b. MAN TALL-αposγ WOMAN TALL-γ iconic-lessβ
‘The woman is less tall than the man.’
Like in more-comparatives, the synthetic form iconic-less cannot be used with adjectives
that belong to the second class as shown by the contrast in (21b) and (22b).
(21) a. ADRIATIC DEEP, AEGEAN LESS
b. *ADRIATIC DEEP AEGEAN DEEP-βiconic-lessα
‘The Aegean sea is less deep than the Adriatic sea.’
(22) a. MAN SMART, WOMAN LESS
b. *MAN SMART WOMAN SMART-αiconic-lessβ
‘The woman is less smart than the man.’
LIS data show a clear connection between loci in signing space and the amount
or degree expressed by gradable adjectives. It bears noting, though, that they do not
mark the exact absolute degree predicated by the adjective as in a 1:1 scale. Rather,
the iconic mapping is proportionally adjusted to the size of the signing space. For
example, LARGE-αposβ can be used to describe both houses and microbes. In these
cases, large dimensions like the size of a house are downscaled, while small dimensions
are magnified, as in the case of microbes.
3.3 Anaphoric properties of visible degrees
In this section, we show that it is possible to anaphorically refer back to the visible/overt
degree by means of a pronoun that points to the locus in which the degree
was previously established. This is shown by the examples in (23)–(25). In (23) and
(24), the pronoun IXβ refers to the degree of GIANNI’s height, while in (24) IXγ refers
to the degree of MARIA’s height. The sentence in (25) shows that the anaphoric relation
can be established even with gradable adjectives denoting abstract properties,
like being cultured.
(23) GIANNI TALL-αposβ IXβ 1 METER 70
‘Gianni is tall. This one (Gianni’s degree) is 1.70 meters.’
(24) MAN TALL-αposβ WOMAN TALL-βiconic-moreγ . IXβ 1 METER 70. IXγ
1 METER 80
‘Maria is taller that Gianni. This one (Gianni’s degree) is 1 meter 70 and that
one is 1 meter 80.’
(25) GIANNI CULTURED-αposβ MARIA CULTURED-βiconic-moreγ .
CULTUREDαposβ CULTUREDαposγ . IXβ IXγ DIFFERENCE A-LOT.
‘Maria is more cultured than Gianni. The difference between this one (Maria
degree of culture) and that one (Gianni’s degree of culture) is a lot.’
LIS signers can also use dual or trial pronouns to refer to two or three degrees
when they are established vertically into two or three different loci as shown by the
sentences in (26a) and (26b):
694 V. Aristodemo, C. Geraci
(26) a. GIANNIa TALL-αposβ MARIAb TALL-βiconic-moreγ . THE-TWOβ−γ
DIFFERENCE A-LOT
‘Maria is taller than Gianni. The difference between these two (Gianni’s
and Maria’s degree of tallness) is a lot.’
b. GIANNIa TALL-αposβ MARIAb TALL-βiconic-moreγ MATTEOc TALLγ
iconic-moreδ. THE-THREEδ−γ −β DIFFERENCE A-LOT
‘Maria is taller than Gianni and Matteo is taller than Maria. The difference
between these three (Gianni’s, Maria’s and Matteo’s degree of
tallness) is a lot.’
Furthermore, once the scale becomes available, any point/degree on the scale can
be used to establish a new locus that becomes an antecedent for a possible anaphoric
relation. This is shown by the example in (27) where it is possible to refer to a degree
between Gianni’s greatest degree of height and Maria’s greatest degree of height. The
representation of the scale is illustrated in the figure in (28) where γ represents the
crucial degree above which passengers must pay the fare.
(27) Scenario: All passengers who are below 1.50 meter travel free on public
buses.
GIANNI TALL-αposβ MARIA TALL-βiconic-moreδ 1 METER 50 IXγ MEANS
MARIA TICKET PAY MUST
‘Maria is taller than Gianni. This point (IXγ ) is 1 meter 50, therefore Maria
has to pay the fare.’
(28)
Notice that in all these cases the pronouns are not referring to the nominal elements.
This is shown by the contrast in (29)–(30). The pronoun in (29) anaphorically refers
to the degree of height of GIANNI and MARIA, therefore it is semantically incompatible
with the adjective BEAUTIFUL. In (30), the pronoun anaphorically refers to the
nominal elements and therefore it is compatible with the adjective BEAUTIFUL.
(29) a. # GIANNIa TALL-αposβ MARIAb TALL-βiconic-moreγ .
THE-TWOγ −β BEAUTIFUL
Intended meaning: ‘Maria is taller than Gianni but both (degrees of
tallness) are beautiful.’
Visible degrees in Italian Sign Language 695
b. # GIANNIa TALL-αposβ MARIAb TALL-βiconic-moreγ MATTEOc
TALL-γ iconic-moreδ. THE-THREEδ−γ −β BEAUTIFUL
Intended meaning: ‘Maria is taller than Gianni and Matteo is taller than
Maria but all the three (degrees of tallness) are beautiful.’
(30) a. GIANNIa TALL-αposβ MARIAb TALL-βiconic-moreγ .
THE-TWOa−b BEAUTIFUL
‘Maria is taller than Gianni but both are beautiful.’
b. GIANNIa TALL-αposβ MARIAb TALL-βiconic-moreγ MATTEOc TALLγ
iconic-moreδ. THE-THREEc−b−a BEAUTIFUL
‘Maria is taller than Gianni and Matteo is taller than Maria but all the
three are beautiful.’
However, if the adjective does not respect the morphological or/and the phonological
requirements, anaphoric relations cannot be established. This is so, because no
location in the signing space has been established as a locus. The relevant example is
provided by the adjective LONG shown (12) and repeated here in (31). Although it is a
Size and Shape Specifier, this adjective does not meet the phonological requirement:
its movement is not perpendicular to the plane of articulation. Hence, the starting and
end points of the sign are not morphologically associated with loci, and they cannot
be overt realizations of degrees.8 Indeed, comparatives with LONG are instantiated by
the analytic form only, as shown by the contrast in (32). The sentence in (33) shows
that a pointing pronoun cannot single out a locus for anaphoric reference.
(31) LONG
(32) a. WORKSHOP LONG SEMINAR MORE
b. *WORKSHOP LONG SEMINAR LONG-αiconic-moreβ
‘The seminar is longer than the workshop.’
(33) *WORKSHOP LONG. IXβ TWO HOUR
Intended meaning: ‘The workshop was long. Its duration (IXβ) was two
hours.’
We suggest that the ungrammaticality of the sentence in (33) reflects what usually
happens in spoken languages, where degrees are not overtly represented. The ungrammaticality
of (33) is comparable to that of the English examples in (7) and (8)
repeated in (34). Crucially, visible iconic degrees allow those readings as shown in
the examples in (23)–(26).
8An anonymous reviewer asked whether forcing the presence of boundaries at the edges of gradable adjectives
makes degree scales visible as an effect of coercion, as reported for ASL (Wilbur et al. 2012:198).
We tested this with some adjectives including THIN and LONG and the degrees are never accessible.
696 V. Aristodemo, C. Geraci
(34) a. # John is taller than Bill. It is 1.70 meter.
Intended meaning: ‘John is taller than Bill and his high is 1.70 meter.’
b. # John is tall. It is 1.70 meter.
Intended meaning: ‘John is tall and his high is 1.70 meter.’
If this is the case then the issue of what counts as ‘overtly visible’ in spoken languages
becomes immediately relevant. We follow Kennedy (1997) in assuming that measurephrases
denote overt degrees. Interestingly, pronominal reference to measure-phrases
is possible in spoken languages too, as shown by the example in (35):
(35) Mary is 2 meters tall. John is taller than that.
Finally, the availability of dual and trial pronouns in degree constructions exemplified
in (26) above is reminiscent of the similar pattern obtained with modal and
temporal anaphora in Schlenker (2013) (see Example (1)). We take this as evidence
suggesting that the same anaphoric system is at work in both domains and in the
nominal domain as well.
4 Iconic scales and scalar analysis
Some gradable adjectives in LIS iconically represent degree variables, which can also
be anaphorically retrieved by pronouns. Specifically, if a gradable adjective meets
both the morphological and phonological requirements illustrated in Sect. 3, an iconic
mapping is realized between the two loci singled out by the movement component
of the sign and semantic degrees. The movement marks two meaningful points in the
signing space (the starting and the end point of the sign). Intuitively, the iconic representation
of the adjective looks very similar to the concept of scale in which points
are ordered along a line in which the starting point corresponds to the standard degree
while the end point corresponds to the reference degree (see Wilbur et al. 2012
for a similar observation for ASL).9 The direction of the movement marks the (positive
or negative) relation between them. Notice that the two points can be marked
in different ways at the morpho-phonological level. Specifically, TALL-αposβ and
DEEP-CL-αposβ iconically map the two points along a vertical line, while BIGαposβ
and CULTURED-αposβ map the two points along a horizontal line (see the examples
in (11) above). In the case of BIG-αposβ the points in space are marked by the
movement of the arm, while in the case of CULTURED the points are identified by the
aperture of the fingers.
Degree variables are realized as loci in the signing space, while the direction of the
movement of the hand marks the degree relation. These two pieces are the building
blocks of the degree morphology of sign language. The visibility of degree variables
can be taken as evidence in favor of the Scalar Analysis. Specifically, we argue that
Loci crucially mark degrees; the facts from LIS thus cannot be immediately explained
9One crucial difference between our proposal and that of Wilbur et al. (2012) is that in our proposal neither
of the two points represents any of the scale boundaries but just two of its degrees along a scale which is
itself visible, see the discussion in Sect. 3.3. Furthermore, Wilbur et al. (2012) do not discuss comparatives
or anaphoric relations.
Visible degrees in Italian Sign Language 697
within the Delineation Semantics approach, where no reference is made to degree
variables.
Following the scalar approach, we provide an explicit definition of iconic degree
scale, which is modeled after the Kennedy and McNally (2005) definition of
scales:10
(36) Iconic degree scale
An iconic scale is the order-preserving mapping of a set of ordered degrees
onto a set of ordered points in the signing space (i.e. a line on the horizontal,
vertical or lateral plane). Each degree of the scale is represented as a point
along a line.
Kennedy (1997) argues that gradable adjectives have three semantic constituents:
a reference value (the degree of the property denoted by the adjective), a standard
value (some other degree to which the reference value is compared to) and a degree
relation (the relation between the reference and the standard value). In comparatives,
the comparative clause (the complement of than) introduces the standard value that
the reference value (introduced by the main clause) is compared to. Degree morphemes
denote relations between the reference value and the standard value. LIS
overtly shows each of these components in both positive and comparative forms: the
standard value is provided by the first locus marked by the movement, the reference
value is marked by the second locus and the degree relation is marked by the directionality
of the movement.
Furthermore, Rullmann (1995) and Kennedy (2001) propose that degrees of
antonymous adjectives map identical arguments (degrees) onto the same scale, but
in the opposite ordering relations. LIS offers iconic evidence for this mapping. The
difference between antonymous adjectives that iconically map degrees is given by the
direction of the movement. This is exemplified by the case of TALL/SHORT in (37)
(see also fn. 7). The only difference between them is the opposite direction of the
movement. This holds for all the adjectives in which degrees are overtly represented.
We take this as evidence of an inverse ordering relation between degrees on the same
scale.
(37) a. TALL b. SHORT
10Kennedy and McNally (2005:349): “Gradable adjectives are those adjectives that map their arguments
onto abstract representations of measurement, or degrees, which are formalized as points or intervals partially
ordered along some dimension (i.e. Tall is ordered by a measure of height). A scale is a set of ordered
degrees.”
698 V. Aristodemo, C. Geraci
5 Conclusion
In this short paper, we showed that LIS gradable adjectives are divided in two groups:
adjectives with iconic degree mapping and adjectives without iconic degree mapping.
We also showed that there are two strategies that instantiate comparative constructions:
the first is expressed by using the analytic forms MORE or LESS, the second is
expressed by using the synthetic forms iconic-more or iconic-less which iconically
map degrees. Gradable adjectives with iconic mapping can use both strategies, gradable
adjectives without iconic mapping can use the analytic form only. The iconic
mapping provides empirical support for a degree-based approach of gradable adjectives
through the fact that loci in signing space are associated with degrees and these
loci become antecedents for later pronouns. To account for these facts, we introduced
iconic scales and iconic degrees as the overt counterparts of scales and degree variables.
An iconic scale is the order-preserving mapping of a set of ordered degrees
onto a set of ordered points in the signing space. Iconic degrees are points along
these lines.
The treatment of gradable adjectives and comparative constructions is an active
debate in the semantic literature. The empirical evidence coming from LIS (and possibly
other sign languages) nicely fits within a scalar approach and raises non-trivial
problems for formal approaches that do not contain degrees as part of the ontology.
In the spirit of reaching both descriptive and explanatory adequacy, our work scores
a significant point in favor of the scalar analysis.
Acknowledgements We wish to thank our LIS consultants Mirko Santoro and Lorenzo Laudo. This
work would not be possible without them. The work also benefited from discussion with Philippe
Schlenker, Jeremy Kuhn, as well as audiences at FEAST 2015, IGG41, CGG25 and Workshop on Gradability
Scale structure and Vagueness: Experimental perspective.
References
Beck, Sigrid. 2011. Comparison construction. In Semantics: An international handbook of natural language
meaning, eds. Claudia Maienborn, Klaus von Heusinger, and Paul Portner. Vol. 2, 1341–1389.
Berlin: de Gruyter.
Brentari, Diane. 1998. A prosodic model of sign language phonology. Cambridge: MIT Press.
Burnett, Heather. 2015. Comparison across domains in Delineation Semantics. Journal of Logic, Language
and Information 24: 233–265.
Cresswell, Max J. 1976. The semantics of degree. In Montague grammar, ed. Barbara Partee, 261–292.
New York: Academic Press.
Demonte, Violeta. 2011. Adjectives. In Semantics: An international handbook of natural language meaning
eds. Claudia Maienborn, Klaus von Heusinger, and Paul Portner Vol. 2, 1314–1340. Berlin: de
Gruyter.
Engdahl, Elisabet. 2001. Versatile parasitic gaps. In Parasitic gaps, eds. Peter Culicover and Paul Postal,
127–145. Cambridge: MIT Press.
Heim, Irene. 1985. Notes on comparatives. Ms., University of Texas.
Heim, Irene. 2001. Degree operators and scope. In Audiatur vox sapientiae, eds. Caroline Féry and Wolfgang
Sternefeld, 214–239. Berlin: Akademie-Verlag.
Kennedy, Christopher. 1997. Projecting the adjective. PhD diss., University of California, Santa Cruz.
Kennedy, Christopher. 2001. Polar opposition and the ontology of ‘degrees’. Linguistics and Philosophy
24: 1.
Kennedy, Christopher. 2007. Vagueness and grammar: Semantics of relative and absolute gradable adjectives.
Linguistics and Philosophy. 30: 1–45.
Visible degrees in Italian Sign Language 699
Kennedy, Christopher, and Louise McNally. 2005. Scale structure and the semantic typology of gradable
predicates. Language 81(2): 345–381.
Klein, Ewan. 1980. A semantics for positive and comparative adjectives. Linguistics and Philosophy 4(1):
1–45.
Levine, Robert, Thomas Hukari, and Michael Calcagno. 2001. Parasitic gaps in English: Some overlooked
cases and their theoretical implications. In Parasitic gaps, eds. Peter Culicover and Paul Postal, 181–
222. Cambridge: MIT Press.
Quine, Willard V. 1948. On what there is. The Review of Metaphysics 2(1): 21–38.
Rullmann, Hotze. 1995. Maximality in the semantics of WH-constructions. PhD diss., University of Massachusetts,
Amherst.
Sandler, Wendy, and Diane Lillo-Martin. 2006. Sign language and linguistic universals. Cambridge: Cambridge
University Press.
Schlenker, Philippe. 2013. Temporal and modal anaphora in Sign Language (ASL). Natural Language and
Linguistic Theory 31(1): 207–234.
von Stechow, Arnim. 1984. Comparing semantic theories of comparison. Journal of Semantics 3: 1–77.
Supalla, Ted. 1982. Structure and acquisition of verbs of motion and location in American Sign Language.
PhD diss., University of California, San Diego.
Wilbur, Ronnie B. 2008. Complex predicates involving events, time, and aspect: Is this why sign languages
look so similar? In Theoretical issues in sign language research, ed. Josep Quer, 217–250. Hamburg:
Signum Press.
Wilbur, Ronnie B., Evguenia Malaia, and Robin A. Shay. 2012. Degree modification and intensification in
American Sign Language adjectives. In 18th Amsterdam Colloquium, ed. Maria Aloni, 92–101. New
York: Springer.