This research was conducted within my position as a research fellow in Italian Sign Language (LIS) at the University of Udine, within the project: “The teaching of Italian Sign Language (LIS): characteristics, diffusion and Methodologies”, scientific director Prof. Paolo Lorusso. The project was funded by the Departmental Strategic Plan 2022-2025, under the name “The words of education: inclusion, languages, spaces, and development” (EDU-ILSS), U-GOV Code: PSD_2022_2025_DILL_RicDip_Fedeli, Rector’s Decree 1562/2022. I wish to thank the deaf colleagues who worked with me over years, for sharing reflections on both signed and spoken language. I am also thankful to the reviewers, for the precious reflections that helped me to improve this article.
1When speaking about “memory” as a general topic, many definitions of the word come up to our mind, according to the approach that is being followed in each moment. From the psychological standpoint, memory refers to the processes involved in encoding, storing, and retrieving information. Through these processes, individuals can recall past experiences, knowledge, or skills for present and future application. Distinct types include sensory, short-term, or long-term memory, according to the amount of time that is necessary for the brain to create it (Britannica [2024]; Camina & Güell [2017]). Neurologically, memory is a biological function dependent on synaptic plasticity and neural network connectivity in brain regions such as the hippocampus, prefrontal cortex, and amygdala. Neurological studies highlight the role of neurochemical pathways and structural changes in neurons, for which the process of turning short-term memory into long-term memory depends on molecular processes involved in information consolidation (Camina & Güell [2017]; Foster [2002]). All of this translates in “memory” as being stored in the brain, in the head or, in any case, in places above the neckline.
2However, many spoken languages also refer to memory as stored in other places in the body, such as the heart or the muscles. Examples of these uses can be found in sentences such as “it is etched on my heart”, in English, or in collocations such as “c’est ancré dans notre coeur et notre mémoire” (“anchored in our heart and memory”), in French. Examples related to “muscle memory” include sentences such as “the general shape of it is still in my muscle memory”. In French, memory seems to relate more frequently to the sensory-motor system, with collocates such as “mémoire visuelle” (“visual memory”), “mémoire auditive” (“hearing memory”), and “mémoire olfactive” (“olfactory memory”). While the first use may be associated to the “heart” in a metaphorical way, as a place where emotional memory is created, “muscle memory” might go beyond metaphors, especially when considering the most recent advancements in neurological studies (Baddeley [1974]; Cann & Ross [1989]; Ackerman [1992]). In fact, as Kukushkin et al. [2024] demonstrated, memory also is stored in non-neural cell lines (thanks to the “spacing effect”, first introduced by Ebbinghaus [1985]), providing scientific bases to other ways to memory creation and interpretation, based on experiences and their repetition. Studies on the interaction between the cognitive neuroscience of bodily self-consciousness (BSC), haptic technologies and robotics are opening the way to a new definition of memory which is related to the sense of self as the accumulation of multisensory knowledge in what is called “body memory” (Repetto & Riva [2023]; Riva [2018]; Riva, Serino, Di Lernia, Pavone & Dakanalis [2017]).
3As research in medicine gives more space to the definition of bodily experience and self-perception, research in linguistics is re-discovering the importance of the body and its experiences in communication and language creation. Further attention is given to the way in which culture is created in triadic interaction, and children acquire language in interaction with their caregivers and the support of objects or gestures, which become part of the communication process as such. In their consensus paper from 2023, Reggin, Gómez Franco, Horchak, Labrecque, Lana, Rio & Vigliocco write
Theories of embodied cognition suggest that modality-specific systems for perception, action and emotion are involved in language processing (Barsalou, 1999; Vigliocco et al., 2009; Wilson, 2002; Zwaan, 2003). While empirical evidence in support of this view has primarily focused on adult cognition (see Horchak et al., 2014, for a review), a growing body of research suggests that children also rely on sensorimotor simulations to acquire vocabulary and develop conceptual knowledge (see Wellsby & Pexman, 2014a, for a review). […] Language acquisition is always embedded in a face-to-face context: there are objects acted upon and talked about, and caregivers providing additional cues, such as gestures, that can bolster language development (Vigliocco et al., 2019) in a manner that can be culturally specific. (Reggin, Gómez Franco, Horchak, Labrecque, Lana, Rio & Vigliocco [2023: 2])
- 1 Although studies in sign language are mostly related to deaf and disability studies, the position h (...)
4Theories in embodied cognition allow the creation of a shared theoretical ground between studies in spoken and signed languages, suggesting the possibility for contents and culture to pass through one or the other modality according to the specific abilities1 of the child. Given the way children are exposed to spoken and gestural stimuli at the same time, specializing in one or the other modality only later in life, the interest of this study is to reflect on the way in which speakers and signers coming from different cultural and linguistic backgrounds represent the meaning for the word “memory”. This word, holding both an abstract and physical meaning, allows us to challenge theories that fit both in embodied cognition as in representation studies.
5In line with the way Borghi, Capirci, Gianfreda & Volterra [2014: 1] defined “abstract word meanings”, the word “memory” could either refer to something that “do not have a single, easily identifiable, imaginable and concrete referent” (the cultural concept of memory), as well as something that is identifiable and concrete, such as the specific areas of the brain where memory is stored. This tension between the abstract and concrete in the meaning of the word “memory” makes it interesting to explore both its spoken and signed manifestation. In the already mentioned article by Borghi, Capirci, Gianfreda & Volterra [2014], different cognitive and representation theories are discussed against the need to explain the way abstract concepts occur in language.
6In this study, we refer to the Conceptual Metaphor Theory (CMT) and its several applications to the study of both spoken and signed languages (Lakoff & Johnson [1980]; Gibbs & Steen [1999]; Boroditsky & Ramscar [2002]; Casasanto & Boroditsky [2008]; Lai & Boroditsky [2013]). Briefly, the CMT states that abstract concepts are related to concrete experience of the world (bodily movement, perceptual interaction, etc.) through image schema, which are cognitive patterns allowing humans to organize actions and behaviors in an efficient way (Johnson [1987]). Used in several scientific domains such as cognitive linguistics (Gallagher [2005]), or neurology (as “body matrix”) (Riva, Serino, Di Lernia, Pavone & Dakanalis [2017]), the image schema is a recurring cognitive structure that emerges from sensory and bodily interactions with the environment. It organizes our experiences and serves as a bridge between concrete physical actions and abstract thought processes (Johnson [1987]). In conjunction with the capacity for conceptual metaphor, which allows human beings to map experiential structure from the imagistic realms of sensory-motor experience to non-imagistic (abstract) ones, image schemas were hypothesized to provide one of the embodied anchors of the entire conceptual system.
7The idea that metaphor is not merely a literary device, but a fundamental mechanism of human thought and action was first suggested by Lakoff and Johnson, in their 1981 paper titled Metaphors we live by. In their work, they state that metaphors are not only figures of speech but are deeply embedded in our conceptual system shaping how we perceive, think, and interact with the world. For example, metaphors like “time is money” influence how we structure our understanding of time as a resource to be spent or saved. The article was the first to refer to “conventional metaphors” as reflecting systematic mapping between conceptual domains, such as understanding “argument” in terms of “war” (e.g., “He attacked my argument” or “I defended my point”). By framing experiences in particular ways, metaphors influence our choices and actions. For example, framing a debate as a “war” might lead to combative attitudes, whereas framing it as a “dance” could encourage collaboration. While many metaphors are universal due to shared human experiences, others are culture-specific, influenced by social practices, values, and beliefs. In all cases, we automatically and unconsciously use them to shape the world we live in.
8In 2005, in his paper “How the body shapes the mind”, Shaun Gallagher brings this idea a little further, discussing the way in which it is the body to shape the way we describe things, with its perceptual system and the way it moves in the world. Gallagher brings the discussion around meanings back to the body itself, as the place where all learning processes begin, and to which we must relate in our ability to express ourselves. In his words:
9Consciousness, as it emerges ontogenetically, consists of a proprioceptive awareness that is already embedded in an ecological framework, and is in intermodal communication with another exteroceptive sense (touch). It is already structured as a pre-reflective, non-conceptual self-consciousness, and involves an experiential differentiation between self and non-self (Gallagher [2005: 106]).
10As a modality of the human body, generated out of movement, language could thus be a translation of embodied spatial frameworks into linguistic forms, which could be gestures. In this framework, spoken language would be “a special kind of oral motility. Speech on this view would be a sophisticated movement of the body” (Gallagher [2005: 107]). What makes a body movement a “speech act” is something that could be easier to uncover when looking into sign language studies and the areas of the body to which signs refer to, technically called “locations”, as well as the way hands and body moves in the “signer’s space”, both considered parameters of any sign language.
11Sarah Taub [2000; 2001] has demonstrated how sign languages blend physical gestures with conceptual metaphors, using the body and space to represent abstract ideas. To explain the way in which ASL constructs metaphors, highlights verb agreement patterns, and explores the layered complexity of ASL poetry, Taub introduces the “analogue-building model”. The image below was taken from her work published in the year 2000, and it depicts the possible process through which the initial concept “tree” is visualized as the sign depicted in (d) through a cognitive process including image selection, schematization and encoding (Figure 1 for our discussion, Figure 3 in her work and all related quotes).
Figure 1. Example of analogue-building model, as in Taub [2000: 35]
12As in Taub’s original discussion,
The analogue-building process begins with a concept that needs a linguistic representation. Such concepts are potentially multi-modal and densely packed with information. For example, the concept “tree” (Figure 3a) probably contains at least visual images of various trees, tactile images of how bark and leaves feel, auditory images (for hearing people) of leaves rustling and branches groaning, kinesthetic images of climbing trees or cutting wood, even images of smells and tastes associated with trees, along with encyclopedic information about the life cycle and uses of trees. Out of this potentially vast amount of information, we select a sensory image to stand for the entire concept. This image must be in a modality that the language can directly represent – e.g., auditory for spoken languages, visual for signed languages, temporal for either one. For ASL TREE, the image is a tree growing out of flat ground, with branches atop a bare trunk (Figure 3b). (Taub [2000: 35])
13The process then continues with the selection of those details that are considered salient to a conceptual/semantic category, in a process that can be called ‘schematization’. In the case of the sign for TREE, the image of a tree growing out of the ground is made of three main components: a flat level surface, a tall vertical shaft emerging from it, and a complex branching structure atop that shaft, as in (c). This stage is cultural-specific, given the way distinct parts of a tree is considered salient to different sign languages (for example, in LIS, the only part of a tree which is selected is the trunk). The final – visible – step is encoding the schematic image into a linguistic form, by selecting articulators that can be arranged in a way that preserves the spatial structure of the original image.
14The reference to a scheme also recalls the “image schemas” mentioned above. These schemas play a fundamental role in how humans comprehend language, reason, and metaphorically interpret the world; thus, they should also be applicable to the way in which sign language users may relate to objects of the world (from now-on, “referents”). Perception and action are the bases for the definition of affordances, as “invitation to act” that are inherent to the shape and features of an object or situation of the world, in each time (Gibson [1979]). As such, affordances could play a key role in defining the parts of the referent that are important to retain, and to represent in the sign. As a preliminary exploration in the direction of understanding the relation between signs and the affordances of their referents in the world, De Monte [2023, 2024] analyzed signs from five different sign languages in relation to affordances suggested by these signs, activating specific image-schema. In her research, De Monte identified several signs that are identical in different sign languages, not necessarily related by any familiarity (as JSL and LIS, for example). The case reported in figure 2 is one of these, showing the sign for UMBRELLA, identical in Italian (LIS), British (BSL), American (ASL), Japanese (JSL), and Brazilian sign language (LIBRAS) (De Monte [2024: 98]).
Figure 2. Depiction of the sign for “umbrella” in LIS, BSL, ASL, JSL, and LIBRAS
15Looking at signs like those analyzed by Taub or De Monte (which are strongly iconic), it is fair to state, in a way, that sign language makes the cognitive process on language, and the body responses, visible.
16The concept of iconicity in sign language has long been at the center of a debate based on the definition of the linguistic features of these visual languages and is currently growing to the point where changes are suggested in the way spoken and signed languages are studied. While iconicity has historically been understood as a direct visual or gestural imitation of meaning, especially visible in sign languages, Cuxac [1993] argued for a more nuanced approach that accounts for cognitive, cultural, and linguistic factors. Sign languages exhibit a multifaceted form of iconicity that extends beyond mere resemblance. Rather than being simple imitations of reality, iconic signs are integrated into structured linguistic systems, governed by grammatical and morphological rules. This idea challenges the classical Saussurean notion of arbitrariness as the primary characteristic of linguistic signs, suggesting instead that iconicity and arbitrariness coexist within natural languages.
17The article further examines how iconicity operates at different linguistic levels, including phonology, morphology, and syntax. It highlights that the perception of iconicity is not universally transparent but is mediated by linguistic competence and cultural familiarity. Some signs, while initially appearing arbitrary, reveal an underlying iconic motivation when analyzed within their linguistic and cognitive context. This perspective challenges the assumption that iconicity is merely a peripheral feature of language and instead positions it as a fundamental aspect of linguistic representation. In the year 2000, Cuxac brings this discussion even further, arguing that the consideration of the peculiarities of such visual (and strongly iconic) languages have a retrospective power over spoken language studies. As a result, he encourages the adoption of an approach that considers modality differences (i.e., through a semiology of the channel) even before undertaking the linguistic description of any language, whether spoken or signed. Though this idea was not new to studies in sign language (William Stokoe already suggested similar conclusions in his 1960 paper), Cuxac is the first to address the areas where such changes could occur.
18He argues that phonetic models developed for spoken languages, which are inherently based on the auditory-vocal modality, prove insufficient for adequately capturing the complexities of a visuo-gestural language such as LSF (Cuxac [2004]). He highlights that the core elements of sign language - hand configuration, movement, spatial organization, and facial expressions - do not function in a manner analogous to phonemes in spoken languages. Instead, they operate within a multimodal system where meaning emerges through a combination of iconicity, spatial grammar, and non-manual markers. The dynamic and three-dimensional nature of these features challenges the linear segmentation principles commonly used in phonetic analysis. Similar conclusions are reached by Pizzuto who argues that the multilinear nature of sign languages challenges the traditional linear representation systems in language corpora, highlighting the need for approaches that are capable to capture their spatial and simultaneous features (Pizzuto [2008]).
19To address these challenges, Cuxac advocates a semantic approach that considers the specificities of the visuo-gestural modality. This perspective moves beyond the constraints of traditional phonetics by integrating the semiotic dimensions of sign languages, particularly their capacity for simultaneity and spatial structuring. He suggests that a refined analytical model, grounded in the particularities of sign languages rather than extrapolated from spoken language theories, is necessary for an accurate representation of LSF’s phonetic organization. Cuxac’s work supports the idea that sign languages utilize a gestural iconicity framework, where linguistic units emerge from both conventionalized and context-dependent representations. His findings suggest that sign language phonology must be studied within a broader semiotic perspective, acknowledging the interaction between formal and semantic structures.
20These are also the conclusions drawn by other sign language linguists such as Corrine Occhino [2016, 2017] who builds on cognitive linguistic theories to argue that phonology and morphology in sign languages exist on a gradient rather than as discrete systems. She examines American Sign Language (ASL) and Brazilian Sign Language (Libras), demonstrating that certain phonological parameters, such as handshape, systematically correlate with semantic categories. For example, the claw handshape is recurrently associated with signs representing concave and convex objects (e.g., BALL, BOWL), reinforcing the idea that phonological form carries inherent meaning. In her research, Occhino undermines the assumption that phonemes in sign languages are purely formal units. Instead, she argues for a usage-based approach, where linguistic structures emerge dynamically from language use and cognitive embodiment. This perspective aligns with the broader view that linguistic categories in sign languages cannot be rigidly separated into phonology and morphology but must be understood as existing within an integrated system. Such conclusions are being supported by more authors, sharing the idea that a semantic approach to sign language description is the leading way to a better understanding of its structure (Garcia [2024]; Thompson, Perlman, Lupyan, Sevcikova Sehyr & Emmorey [2020]).
- 2 Sketch Engine is a corpus analysis tool designed for linguistic research, lexicography, and languag (...)
21Given the semantics of the word “memory”, the contribution of psychology and neurology to its definition in terms of image schema and body activation, the role of the body in defining speech and of metaphors in describing the world, and the semantic approach to sign language studies, how does sign language relate to these concepts? Will it relate to the words and meanings used in the corresponding spoken language or will it relate to image-schemas and deeper cognitive processes related to perception and action, and a more iconic, visual-based organization of the world? To provide a preliminary answer to these questions, an analysis on spoken language corpora of spoken English and French has been conducted to isolate their meaning among other interpretations of the word. The analysis has considered the corpora stored on Sketch Engine2, starting from the word “memory” for English, and “mémoire” for French.
22The research in English returned more than 6 million occurrences of the word “memory” as noun, in the “English Trends (2014-today)” corpus. The thesaurus linked the word to 6 lemmas having a similarity score of over 0.400 points (that is, the percentage of collocates the synonym has in common with the searched word): “picture”, “thought”, “sense”, “image”, “past”, “love”. The search through modifiers of the word mentions “muscle” as in “muscle memory”, which is among the most used modifiers in the corpus. Thus, the preliminary English search related the word “memory” to images, feelings, and time. In the case of the French “mémoire”, the results from the “French Web 2023 (frTenTen23)” corpus brought up 28,200 instances of the word, correlating it to sensory-motor memory, as reported in the introduction to this paper. The highest similarity score (more than 0.080) is observed for words such as: “ram” (RAM), “ame” (soul), “ecran” (screen), “presence” (presence), “mere” (mother), “année” (year). The French use of the word is mostly related to electronic devices, time, and family, with some reference to the spiritual dimension of the word, as in the use of “ame” as the place where memory is built.
- 3 Among the tools made available by SketchEngine, a “word sketch” is a summarized grammatical and col (...)
23The word sketch3 has been then used to refine the research on different sign languages, which has been conducted using online and paper dictionaries in Italian, British, American, Japanese, and French sign languages (LIS, BSL, ASL, JSL, and LSF, respectively). The interest for these specific varieties is justified by the physical and cultural distance between the European variations (LIS, BSL and LSF), and the American and Asian varieties. Online versions of the signs were compared with papers ones and vice-versa, in search for consistency. The signs reported in the example below (Figure 3) are all taken from Spreadthesign.com, an online glossary of signs hosting over 500,000 “suggestions of signs from different sign languages around the world”. The website hosts 44 spoken and signed languages, of which 5 are English variants (from Australia, India, New Zealand, United Kingdom, and United States).
24The search for the word “memory” in the online glossary resulted in 36 variations of signs for the word, all relating to the forehead as their location. In the five variants being observed, the search has produced the following results:
Figure 3. Sign for “memory” in LIS, BSL, ASL, JSL and LSF. The text below the images describes the movement on the hand by the signer.
|
LIS
|
|
|
|
Double-tap movement on the forehead
|
|
BSL
|
|
|
|
Double movement, head nodding
|
|
ASL
|
|
|
|
Outwards movement, hand closed while moving
|
|
JSL
|
|
|
|
Inwards movement, hand closes while moving
|
|
LSF
|
|
|
|
Thumb moves from forehead to non-dominant hand
|
- 4 CLARIN (standing for Common Language Resources and Technology Infrastructure) is an European Resear (...)
25As noted in the movement description (reported in the line below the signs in Figure 3), changes between signs would mostly appear in the selection of the handshape, or in the movement of the hand with respect to the forehead (outwards vs inwards). As for ASL, another variant of the sign (not reported in figure 3) is identical to the one for LSF. In fewer occasions, the sign was completed by an additional productive sign, related to a more complex meaning for the word (such as ‘memory of the past’). Productive signs are strongly related to context (Branchini & Mantovan [2022]), and it has therefore not been analyzed any further as it is out of the control of the researcher. Corpora in sign language, the largest of which is hosted by CLARIN4, were not considered for this part of the study, as some of the languages analyzed here are not present in the collection.
26Sign language has long been studied, and described, in relation to spoken language. Yet, when observing the profound distance between the spoken meanings attached to the word “memory” in English and French and the way in which the same meaning is channeled in different sign languages of the world, the distance between the two becomes evident. Sign language seems to relate to the human experience of memory much more than it relates to other meanings attached to these words in the spoken language. Although the analysis provided is strongly limited by the difference in the amount and type of data observed, it does provide evidence to the fact that, for sign language, memory is stored in the head. As it can be seen in the example provided in Figure 3, the English word for “memory”, and its translation to the Italian “memoria”, and the French “mémoire”, are all represented with signs referring to the head as its preferred location. In fact, all the signs seen above, the head is the place where the “memory” is ‘sticked’ (in LIS and LSF), ‘grasped and stored’ (in BSL and JSL), or ‘held’ (ASL). In other words, the analogue for the meaning of the word “memory” is something that can be grasped, and manipulated, to become one with the signer. Similar constructions, regarding the sign for [REFLECHIR] (to think / reflect on), were defined by Cuxac, 2004, as ‘condensed conceptual metaphors’, signs that are built to represent meaning through the iconic and semantic features of their parameters.
27Interestingly, research for the meaning of the word “tradition,” as a collective memory, results in signers from 30 countries producing remarkably similar signs referring to something cyclically moving from past to present or within the present (Figure 4).
Figure 4. Sign for “tradition” in Italian (LIS) and French (LSF) sign languages
|
LIS
|
|
|
|
Hand moving cyclically from dominant shoulder to neutral space
|
|
LSF
|
|
|
|
Hand moving cyclically from dominant shoulder to neutral space
|
- 5 Should the handshape represent and action such as “holding something” or an “entity/situation”, dep (...)
28In the figure presented above, the Italian and French signed version for “tradition” are identical, except for the handshape for both hands. In the case of LIS, the handshape suggests something that is being “held,” or an “entity or situation”5 recurring in an analogous way over time. In LSF, the “entity” is referred to through pointing, as the handshape suggests. In both cases, the movement of both hands from the shoulder towards the center of the signer’s space is performed by alternating hand positions in small circles, iconically representing repetition. Similarly, the forward movement of the hands is correlated with the experience that signers of these languages have of time and space, as something that moves forward, which is the underlying iconic motivation that the sign carries when analyzed within their linguistic and cognitive context, in a way as suggested by Cuxac [1993] and recalled in section 3.
29In the light of the theories discussed above, the head-related meaning for “memory” seems to imply a stronger relation of the meaning of signs to the experience of the body. Similarly, tradition and collective memory seem to relate to cyclic repetitions, exactly as it would be described from an experiential point of view. According to these ideas, also introduced in sections 2 and 3, the signs used in every sign language much relate on body experiences shared among humans, which not always relate to similar descriptions provided by the language spoken by the majority. This difference in reference could be related not only to the linguistic structure of sign language (in terms of manual and non-manual components, use of space, etc.) but also to the result of a body conception as a multitude of meanings (Courtin [2000], Courtin & Melot [2005]).
30In the body, the visual-spatial and signed modality influences the construction of the lexical sign, in a way that remains strongly iconic and visual. If validated by further studies, this understanding of the process of signification underlying the production of signs may represent an important turning point in explaining how different language modalities describe the world that we live by.
31In this position paper, built upon a multidisciplinary approach based on corpus-based research and cognitive linguistic approaches, the individual and cultural concept of “memory” has been discussed. Psychologically and neurologically, memory is rooted in intricate processes of encoding, storing, and retrieving information, primarily associated with the brain’s neural architecture. However, the metaphorical and embodied representations of memory across cultures and languages highlight a broader perspective. For instance, references to “muscle memory” or “memories etched in the heart” highlight how physical and emotional experiences are intricately woven into our conceptualization of memory.
32Sign languages offer a unique perspective on this topic, often reflecting image schemas and embodied cognition principles. By grounding abstract ideas in physical gestures and iconic representations, sign languages emphasize the role of the body in linguistic expression and meaning making. The consistent association of memory with the head in sign languages, as evidenced by the analysis presented here, aligns with psychological and neurological models, yet retains distinct cultural and visual interpretations.
33As research continues to explore the intersections of memory, metaphor, and embodied cognition, it becomes increasingly evident that memory transcends a purely cognitive or biological perspective. Instead, it integrates our sensory, emotional, and cultural experiences, shaping the way we communicate, interpret, and interact with the world. This evolving understanding underscores the need for interdisciplinary approaches to unravel the complexities of memory and its diverse manifestations across language modalities and cultures.