Introduction
Twenty years ago, when work on the intersection of language and dementia was in its relative infancy, Bayles and Kaszniak (Reference Bayles and Kaszniak1987, p. 175) recommended that systematic study of the conversational abilities of dementia patients supplement more typical research within the experimental paradigm, not only because the loss of such abilities was thought to be an early marker of dementia, but also because such studies could help researchers find out how patients used language in everyday life.
The past two decades have seen an energetic and creative response to that call, although, given its importance, perhaps to a lesser degree than one might expect. The majority of work on language and dementia continues to be carried out from a neurolinguistic or psycholinguistic perspective (see Kempler & Goral, Reference Kempler and Goral2008, this issue). To my knowledge, the publication of my revised doctoral dissertation, Conversations with an Alzheimer's Patient: An Interactional Sociolinguistic Study, was the first in the wave of studies to depart from the dominant clinical paradigm with its experimentally elicited language data and to focus on social interaction. In Hamilton (Reference Hamilton1994a), I analyzed the language of open-ended, naturally occurring, and meaningful conversations I had with an individual with Alzheimer's disease over 4.5 years in her nursing home. This study was an attempt to understand not only how communicative abilities and disabilities were related to each other and how they changed over time, but also, importantly, how these were influenced by both preemptive and reactive communicative behaviors on the part of the conversational partner. Ramanathan (Reference Ramanathan1995, Reference Ramanathan1997) and Ramanathan-Abbott (Reference Ramanathan-Abbott1995) then extended this focus on interactional effects to examine how narratives told by speakers with Alzheimer's disease were shaped according to differences in audience and setting.
In subsequent work, I incorporated notions of face (Goffman, Reference Goffman1967), linguistic politeness (Brown & Levinson, Reference Brown and Levinson1987), and positioning theory (Davies & Harre, Reference Davies and Harre1990) to investigate the social construction of identities in oral (Hamilton, Reference Hamilton1996, Reference Hamilton2008) and written (Hamilton, Reference Hamilton, Peyton, Griffin, Wolfram and Fasold2000) communication by individuals with Alzheimer's disease, linking this work to studies that discuss the health consequences of communicative practices involving older adults (see Giles, Coupland, & Wiemann, Reference Giles, Coupland and Wiemann1990; Ryan, Giles, Bartolucci, & Henwood, Reference Ryan, Giles, Bartolucci and Henwood1986). In all of this work, Alzheimer's disease is approached as a human issue within multiple linguistic and social contexts, rather than exclusively as a slowly progressive disease of the brain.
Focusing attention on the most recent scholarship on language and dementia, note that the motivation for sociolinguistic research comes from what Leibing (Reference Leibing, Leibing and Cohen2006, p. 242 ff.) calls the “personhood movement” in Alzheimer's studies (cf. Kitwood, Reference Kitwood1997), in which “personhood” refers to “the person within—the reflexive, immaterial, communicable essence of a person that is located deep within the body, but that is sometimes veiled by symptoms” (Leibing, p. 243). This shift away from the more pervasive medical model of dementia is reflected in its emphasis on the “capacities of the feeling person and not only on his or her losses” (Leibing, p. 255), as well as on a redefinition of memory as “interactive and not individualized” (Leibing, p. 255). Not surprisingly, this shift toward a focus on personhood has been accompanied by a heightened interest in applying the findings of basic research to help individuals with dementia and those who care for them, for example, by enhancing communication and raising self-esteem. In fact, it could be argued that most studies reviewed here were motivated by the observation that “relatively little can be done to arrest the underlying brain disease, [but] much can be done to promote health and wellbeing” (Downs, Clare, & Mackenzie, Reference Downs, Clare, Mackenzie, Hughes, Louw and Sabat2006, p. 248), resulting in a focus on the identification of active coping strategies and the enhancement of social environments for the individual with dementia.
To differentiate the scope of literature reviewed here from that reviewed by Kempler and Goral (Reference Kempler and Goral2008) in this issue, I focus on studies that (1) examine naturally occurring language (i.e., not language produced within clinical or experimental contexts in response to standardized instruments) with attention to variation within that language that relates to speakers’ cognitive or sociocultural attributes (e.g., severity of the disease, first or second language, race, ethnicity, socioeconomic status, and sex) or to contextual dimensions of the language (e.g., interlocutors, physical setting, and social activity); (2) provide a rich understanding of speakers with dementia or contexts of communication that can illuminate sociolinguistic analyses; or (3) call for sociolinguistic investigations to clarify or provide additional evidence for a particular finding outside of linguistics.
This review is structured from the global to the local levels; that is, I begin with literature that frames the study of language and dementia from the societal or cultural perspective, focusing on public discourse surrounding dementia and Alzheimer's disease. I then focus in on contextual dimensions of specific interactions in society, first examining the influences of the type of activity in which participants are engaging and then turning to the shaping influence of the physical environment. The following section zeroes in on aspects of the communication itself, discussing findings related to verbal communication before turning to those related to nonverbal communication. The review closes with implications for future research into sociolinguistic aspects of language and dementia.
Public Discourse: Framing of Dementia and Consequences for Interaction
Snyder (Reference Snyder, Hughes, Louw and Sabat2006) reported the following exchange within a support group meeting for individuals with Alzheimer's disease.
L: Once you have that label [Alzheimer's], it changes the way people communicate with you. Everyone treats you with ..uh.. you know, they don't want to. . .
K: Upset you.
L: Yeah, upset you or anything like that. You're just treated differently. (p. 267)
Why would it be the case that people communicate differently with individuals who have “that label” of Alzheimer's disease? Johnstone (Reference Johnstone2002, p. 227) proposed an answer when she highlighted the relationship between what speakers perceive to be “real, natural or true about the world” and their discourse practices, exploring how elements of ideology “shape what people do with talk, what they do not do, how they interpret discourse.”
What kinds of ideologies may be at work here? I have already mentioned differences in perspective within the scholarly community working on language and dementia. In this sense, Cohen (Reference Cohen, Leibing and Cohen2006, pp. 7–8) juxtaposed what he termed the “total biologization of senility” (where internal disease processes inform all discussions of changes in late life) with the “personhood” turns in clinical practice and social science research mentioned earlier. But clearly I am not concerned solely with scholarly debates; the general public is immersed in public discourse that shapes its views of people with dementia—and this public discourse is indeed what can vary across cultures, societies, and institutions.
Downs et al. (Reference Downs, Clare, Mackenzie, Hughes, Louw and Sabat2006) outlined four explanatory models or belief systems about the cause of dementia that circulate in the public sphere, each of which has implications for the individual with dementia as well as for an understanding of the therapeutic efforts required to provide support. Of these belief systems, Downs et al. (p. 244) claimed that the “neuropsychiatric” model, in which the focus is on the underlying brain disease, dominates Western discussions (but see Ridge, Makoni, & Ridge, Reference Ridge, Makoni, Ridge, Makoni and Meinhof2003, for a South African perspective). In this model, psychosocial aspects of dementia are neglected because “intellect and reasoning are valued above relational and aesthetic aspects.” This connects very clearly to Post's (Reference Post1998, p. 72) notion of the “hypercognitive culture” (cited in Sabat, Reference Sabat2001, p. 320), which values cognition over other dimensions of life. Post maintained that hypercognitive cultures diminish individuals with dementia by “reducing their moral status or by neglecting the emotional, relational, aesthetic, and spiritual aspects of well being that are open to [these individuals], even in the advanced stages of the disease.”
In their useful, multifaceted textbook, Approaches to Discourse Analysis in Dementia, Guendouzi and Muller (Reference Guendouzi and Muller2006, p. 146 ff.) highlighted the importance of critical discourse analysis (cf. Fairclough Reference Fairclough1989, Reference Fairclough1995) in examining the linguistic choices made within media discourse related to dementia. They claimed that such analyses of Western cultural texts reveal “hegemonic discourse practices” that represent dementia as “loss of self” and the individual with dementia as an “empty shell” (see also Kontos, Reference Kontos, Leibing and Cohen2006, p. 195). Guendouzi and Muller stated that these public discourses find their way into language used by individuals in Alzheimer's support groups (and more generally as well).
Snyder (Reference Snyder, Hughes, Louw and Sabat2006, p. 268) made the explicit connection between negative media images and effects on person-to-person communication, stating that “socially constructed fictions such as ‘the loss of the self,’ ‘empty shell,’ and the prospect of ‘a long goodbye’” can affect communication in both subtle and dramatic ways. She introduced Vaihinger's “as if” philosophy (Vaihinger, Reference Vaihinger1924) into her argument that conversational partners can choose between communicating with people with dementia “as if” their self is diminished or “as if” there is still a “self trying to be recognized.” In the former case, it is less likely that interlocutors will be able to “hear the intact messages that are expressions of . . . personal identity.”
Sabat (Reference Sabat, Hughes, Louw and Sabat2006, p. 297) made a similar point from the perspective of positioning theory (Davies & Harre, Reference Davies and Harre1990). In what Sabat termed “malignant positioning,” an individual's actions are attributed to dementia and not seen as possible intentional acts on the part of the person (e.g., when the source of an unwillingness to participate in an assisted living center's bingo game is not understood as possibly being related to an individual's lifelong dislike of bingo). Sabat argued that such malignant positioning can be counteracted when conversational partners see the person with dementia as “much more than a patient—as a person who can have a variety of valued social identities.” The linguistic details of such moment-to-moment shifts in positioning can be found in Hamilton (Reference Hamilton1996).
In fact, it is this critical two-way relationship between interlocutors’ conversational actions (e.g., positioning, communicating “as if,” etc.) and communicative behaviors and levels of self-esteem of the individual with dementia that is at the heart of the sociolinguistic perspective on this issue. When the experience of living with dementia is understood as the result of dialectical interplay between brain disease and psychosocial factors (Downs et al., Reference Downs, Clare, Mackenzie, Hughes, Louw and Sabat2006), it becomes important to extend outward beyond specific analyses of individuals’ linguistic production and comprehension to encompass relevant aspects of the social context of this language use. In the following two sections, I borrow the term “talk-in-interaction” from the field of conversation analysis (cf. ten Have, Reference ten Have1999) to focus attention on the fact that communication can best be understood in relation to its surrounding interactional contexts. I begin by highlighting the “interaction” part of the term, first discussing the activity of which the talk is a part followed by aspects of the physical environment of such talk. I then turn to the “talk” part of the term, discussing verbal aspects of communication followed by nonverbal aspects.
Talk-In-Interaction: Focus on Interactional Contexts
Over the years, studies of language and dementia have typically examined language elicited by standardized testing within a clinical setting or as part of interviews or conversations with a researcher. Because differences inherent in interactional contexts have been shown to result in differences in language produced and comprehended, some researchers have identified different contextual features as being at least partially responsible for contradictory findings across studies (see Light, Reference Light, Blanken, Dittmann, Grimm, Marshall and Wallesch1993; Melvold, Au, Obler, & Albert, Reference Melvold, Au, Obler, Albert, Albert and Knoefel1994). McLean (Reference McLean, Leibing and Cohen2006, pp. 167–168), for example, suggested that some discouraging assessments may result from the artificial experimental conditions under which language was produced as well as from the power differential this creates between subject and researcher. In addition to these contextual differences related to activity and purpose, Ramanathan (Reference Ramanathan1997) identified the relationship between conversational partners as an influential factor on language used by individuals with dementia.
In an attempt to provide interested sociolinguists with sufficient opportunities to study similarities and differences in language used by individuals with dementia across different contexts, I have recommended (cf. Hamilton Reference Hamilton and Davis2005, pp. 237–238) the systematic recording of interactions that go beyond the typical extended conversations and interviews to include those with a wider range of purposes and interlocutors. To that end, some recent work has focused on activities in which individuals with dementia are engaged with interlocutors other than researchers or clinicians. Additional work has problematized aspects of the physical context of the communication, including both the built environment and the function of inanimate objects. These two areas of the interactional context are reviewed in turn.
Focus on Activity
In their longitudinal study of elderly in Taiwan, Glei et al. (Reference Glei, Landa, Goldman, Chuang, Rodriguez and Weinstein2005) found that voluntary participation in social activities (such as playing games; doing volunteer work; and participating in religious groups, business associations, or political groups) has a greater impact on the maintenance of cognitive function than do regular interactions with family members and friends. In a response to this study, Gallacher, Bayer, and Ben-Shlomo (Reference Gallacher, Bayer and Ben-Shlomo2005, p.1) called for more detailed investigations that would “unpack the social engagement ‘black box’” and identify the relative importance of various characteristics of these social activities. Although a great deal more work needs to be done in this area, several studies suggest relevant dimensions.
Not surprisingly, perhaps, activities examined in studies of dementia frequently relate in some way to memory. Some activities for individuals in the relatively early stages of dementia are designed with an explicit focus on memory, either through conversations encouraging personal reminiscence or in the form of memory games. Basting (Reference Basting, Leibing and Cohen2006, p.193) suggested that this encouragement to remember the past may be triggered by assumptions on the part of caregivers that people with dementia have “no meaningful present.” In her study of a day care center for Swedish speakers in Finland, Lindholm (in press) discussed face-threatening aspects of the moment-to-moment management of what she called the “interactional noncompetence” displayed within the context of a memory game based on Swedish proverbs. During the game, a nurse read the first part of a proverb aloud, creating a slot that the individual with dementia was supposed to fill by producing the second part of the proverb. The transcript excerpts of these interactions suggest that the individuals viewed this activity as less of a game than a memory test.
In contrast, Snyder (Reference Snyder, Hughes, Louw and Sabat2006), Basting (Reference Basting, Leibing and Cohen2006), and Kontos (Reference Kontos, Leibing and Cohen2006) all discussed activities that downplay the focus on memory, since “individuals’ own self esteem can be battered within therapies that focus on reminiscence or reality orientation, or even in personal interactions with family members and friends who insist on memory work” (Kontos). For example, in relatively early stages of dementia, when individuals affected can recognize a gap between what they can remember and what they should be able to remember, support groups can be an important activity to counteract attempts by personal or professional caregivers to “maintain positions grounded in ‘reality’” (Snyder, p. 266). Support groups that focus on coping strategies, finding meaningful activities, and struggling with change and autonomy allow “affirming, engaged discourse that is rarely afforded people with dementia as they attempt to navigate mainstream life.” Within this activity, speakers are free from judgmental responses by conversational partners who may be frustrated with what they view as fictional or repetitious talk coming from those with dementia.
For individuals in the moderate stages of dementia, the language and memory challenges associated with reminiscence, proverb games, and support groups may be too great, leading to frustration and shame. Basting (Reference Basting, Leibing and Cohen2006) described the TimeSlips Project, a program she developed in response to this dilemma. This project, which shifts focus from memories of the past to creativity in the here and now, can “open up channels of communication with people with ADRD [Alzheimer's disease and related disorders] and offer both client and caregiver the potential for growth”(Basting, p. 194). Rather than expecting individuals to come up with stories from their past, facilitators hand out a visual image to each member of a small group and ask many open-ended questions of group members, with the resulting creative story being based on all members’ responses. This process provides an “avenue for self-expression that frees [individuals with dementia] from the demands of memory and rational language” (Basting, p. 193).
Finally, when the stage of dementia is even more advanced, Kontos (Reference Kontos, Leibing and Cohen2006) emphasized the importance of routine activities in helping to maintain a sense of normalcy in life. In her ethnographic study of an Alzheimer's support unit of an Orthodox Jewish home for the aged in Canada, Kontos explored the notion of selfhood in the face of what she called “severe cognitive impairment.” She beautifully illustrated “the body's concrete, spatial, and prereflective directedness toward the lived world” (p. 203) in a series of powerful and poignant vignettes that understand automatically performed activities in the nursing home (e.g., Jewish rituals) as “active presence of the past in the body itself” (p. 209). From this perspective, the construction of self need not be reliant on language per se, but can be “enacted in the actual movements of the body.” Kontos argued, furthermore, that this embodied selfhood goes beyond historical rituals to include residents’ mastery of their social world through “the rhythm of ongoing interactions of daily life.”
Focus on Physical Context
In addition to the focus on the shaping contextual influence of the activity type, other scholars have pointed to the relevance of the physical context, including the built environment of, for example, nursing homes and assisted living centers, as well as inanimate objects in the physical surroundings. In an intriguing study of seven Alzheimer's special care units, Zeisel et al. (Reference Zeisel, Silverstein, Hyde, Levkoff, Lawton and Holmes2003) found associations between behavioral health measures and particular environmental design features of those units. For example, as reported by nurse observers, individuals with Alzheimer's disease displayed less anxiety and aggression in special care units that appeared more residential and less institutional and allowed for more privacy and personalization of rooms. Importantly, the amount of variability among common spaces in a facility was negatively correlated with social withdrawal, the variable that is centered most on communication; that is, residents with Alzheimer's disease were observed to initiate more interactions, respond more to social contacts, pay more attention to things around them, and help other residents more when the common spaces of the facility did not all look the same. From an ethnographic approach, Chatterji (Reference Chatterji, Leibing and Cohen2006, p. 227) concurred with Zeisel et al. about the importance of the lived environment, although given her ethnographic perspective, she understood decisions regarding public and private spaces as being related to cultural values (such as autonomy and independence) that are central to the constitution of self and to participation in social life.
As far as I know, no sociolinguistic studies per se focus on the interrelationship of physical context and language (but see Scollon & Scollon, Reference Scollon and Scollon2003, for a useful theoretical framework) used by individuals with dementia, although studies from neighboring fields make the case that issues of place and space are important areas for future linguistic analyses. Other motivation for such studies comes from early examinations of individuals with dementia that identified what some scholars term “context-boundedness” (Appell, Kertesz, & Fisman, Reference Appell, Kertesz and Fisman1982; Obler, Reference Obler1981), an individual's reduced ability to free him- or herself cognitively from the immediate temporal and spatial context. In Hamilton (Reference Hamilton2008) this context-boundedness was evidenced by the vast majority (85%) of communication that referred not to memories of displaced times and places or to abstract thoughts but to people, objects, and actions within the immediate physical environment of the conversation. The balance of talk that did refer to the past was clearly triggered by inanimate objects in the here and now.
Following Kitwood's (Reference Kitwood1990, Reference Kitwood1997) argument that memory should not be understood as inhering solely in individuals, Leibing (Reference Leibing, Leibing and Cohen2006) raised the possibility of external extensions to memory, both in other persons, where family members and friends can become part of an extended or interpersonal memory (see also Ramanathan, Reference Ramanathan1997, on scaffolding) and in inanimate objects that become humanized in their integration into personhood. The importance of greeting cards, photographs, and other such personal holders of memory cannot be overestimated in their role as conversational triggers (see also McLean, Reference McLean, Leibing and Cohen2006, p. 163). Such objects can also assist in constructions of self-identity, as observed by Oppenheimer (Reference Oppenheimer, Hughes, Louw and Sabat2006, p. 202): “When a person can no longer describe to the visiting psychiatrist who they are or have been, their home—its atmosphere and the objects in it—can help to do so.”
Talk-In-Interaction: Focus on Communication
This section focuses on the tip of the inverted triangle, that is, on the communication—both verbal and nonverbal—that occurs within the some kind of interactional context (third section) and is potentially shaped by public discourse surrounding dementia (fourth section). In Hamilton (Reference Hamilton and Davis2005), I used the metaphors of the prism, the soliloquy, the couch, and the dance to highlight four prevalent approaches to the analysis of communication as it relates to dementia. To summarize, the prism symbolizes the focus of some researchers on the disembodied display of a variety of linguistic and communicative phenomena that characterize the breakdown that typically occurs as the dementia progresses. From this perspective, selected linguistic or discursive phenomena are analyzed apart from other linguistic or social phenomena, with no specification of a relationship to the speaker or the context of talk beyond possible subsequent straightforward correlations. The soliloquy represents work that focuses its analytical attention on the investigation of how individuals with dementia integrate the individual linguistic and communicative phenomena identified by the prism approach into a whole discourse. In this approach, linguistic phenomena are analyzed not as separate phonological, morphological, syntactic, semantic, pragmatic, and lexical phenomena, but instead are examined in light of the larger discourse in which they occur—and which they help to construct.
The couch—referring to the stereotypical couch found in a psychiatrist's office—represents work that focuses on personal meaning-making, sense-making, and identity construction carried out by individuals with dementia. This approach examines ways in which language is used to display the (potentially) changing identity of the person with dementia or, alternatively, ways in which breakdowns in language over time are responded to by the individual in terms of meaning and self. And, finally, the dance represents work that focuses its attention on the interaction between language users and the ways in which interlocutors influence each other—both socially and linguistically. In this approach, which highlights the interactions as process rather than as product, researchers examine moment-to-moment turns or moves that display interlocutors’ meaning-making and relationship building as these emerge across the interaction. This is crucially different from the discourse approach characterized by the soliloquy, which focuses exclusively on the language produced by the individual with Alzheimer's disease. In the dance, the interdependence between or among partners is accentuated.
Focus on Verbal Communication
With reference to the metaphors in the preceding section, the dance is represented by Ryan, Spykerman, and Anas (Reference Ryan, Spykerman, Anas and Davis2005) and Pope and Ripich (Reference Pope, Ripich and Davis2005). Ryan et al. examined 50 audiorecorded and transcribed conversations spanning 4 years with a single individual with dementia who lived in nursing home special care unit. Because these conversations were all held with experienced communicators whose sole purpose was to have a conversation, and not, for example, to carry out other nursing home–related tasks, Ryan et al. aimed to characterize “conversation as care” by identifying transcript examples that promoted personhood in Kitwood's (Reference Kitwood1990, Reference Kitwood1997) sense, where the conversational partner continued to show interest in the person and his life story, preferences, emotions, and needs. Ryan et al. revealed strategies that could be used by healthy conversational partners to facilitate conversations with individuals affected by dementia, such as providing proper nouns (rather than pronouns) and longer pauses to allow formulation of responses. The importance of being able to participate in successful conversations is emphasized by the term, the “warmth of reciprocity,” used by Ryan et al. (p. 33) to describe these activities.
Working in the much underexplored area at the intersection of race, gender, and dementia, Pope and Ripich (Reference Pope, Ripich and Davis2005) provided provocative observations regarding topic introduction and maintenance, as well as the impact of institutional speaking practices that may limit the potential for social interaction or response to cues based on common gender or ethnicity. Although they were able to examine only same and differing gender pairs (individual with dementia and caregiver or interviewer) within monoracial African American and European American encounters, contrasts in permissible topics and speaking strategies strongly suggest the importance of further investigations in this area, including expansion into interracial communication—both within groups of individuals with dementia and between caregivers and those receiving their care.
The soliloquy is represented by Hamilton (Reference Hamilton2008), in a study that examined the intersection of narrative, identity, and memory in five tape-recorded conversations over 4.5 years involving an individual with moderate to severe dementia. In this work, I focused first on a set of 204 clauses that referred to the past, differentiating clauses that were part of conversational narratives (56, or 27%) from independent clauses I termed “narrative traces” (148, or 73%). I then identified and examined the linguistic construction of the storyworlds in 15 short conversational narratives comprising the 56 narrative clauses, including reference to people, objects, activities, states, time, and space. Frequency lists of lexical items and phrases provided evidence that the abilities to use these linguistic building blocks to construct vivid, meaningful storyworlds were highly diminished. These problems began with relatively infrequent references and were exacerbated by the weak semantic nature of those references that were actually made (overwhelmingly including large numbers of personal pronouns and lexical items such as people, part, thing, place, do, get, make, put, then, when, in, on, and over). I argued that these problems had consequences not only for the coherence of the narratives as texts but also for the discursive construction of the narrator's identity as well.
The prism approach is represented by Maclagan and Mason (Reference Maclagan, Mason and Davis2005) in their analyses of conversational cohesion, lexical richness, and word class within 50 spontaneous conversations involving one institutionalized individual over 4 years with four different interviewers. The researchers expected that word class and lexical richness would decline over time, but that many features of cohesion would not. Counter to expectations, both the word class analysis (nouns, pronouns, verbs, and adjectives) and the lexical richness showed “fairly wide swings between interviews,” indicating that communicatively the speaker appeared to have “good days and bad ones” rather than a “downhill slide through time” (p. 158). The authors argued that the large day-to-day variability within the language of a single person “highlights the difficulties involved in obtaining valid analyses of the language of subjects with Alzheimer's disease” (p. 163) and should cause researchers to interpret results with great caution.
This prism approach is also represented by March, Wales, and Pattison2 (2006) and Davis and Bernstein (Reference Davis2005) in their very different examinations of the “empty” speech also identified in the preceding studies. In their experimental investigation of spatial deixis (e.g., here, this), person deixis (e.g., I, he), and noun use by 26 speakers with mild to moderate Alzheimer's disease and 26 demographically matched healthy elderly in response to four discourse tasks (description of a single picture, narrative description of sequential pictures, cartoon, and route description in shared vs. unshared contexts), March et al. postulated that speakers with Alzheimer's disease would rely on the communicative context more often (i.e., producing fewer nouns but higher rates of deixis) than the healthy elderly in their attempts to compensate for specific difficulties in naming. They found that the Alzheimer's speakers produced significantly fewer nouns only in the single picture task, used significantly more inappropriate spatial deixis when the context was not shared between speakers (unshared map task), and showed a reduced person deictic use in the unshared map task and in the narrative description of sequential pictures. As a result of these findings, March et al. called for the inclusion of multiple tasks in any future study of discourse production related to dementia; future examinations of the relationship between the severity of Alzheimer's disease and the use of nouns and deixis within less structured discourse types, such as conversation; and future examinations that focus on the perspective-taking ability in deictic use.
By investigating the multiple uses of lexical items thing and that within spontaneous conversations over the course of 2 years with a single speaker at the moderate to moderately severe stage of dementia, Davis and Bernstein (Reference Davis, Bernstein and Davis2005) argued for the importance of context in any analysis of “empty” language. Their instructive fine-grained analysis shows that the word thing was used either to substitute for a discrete, countable, and currently visible object or in a colloquial and formulaic phrase (e.g., There ain't no such thing) used for emphasis or humor. Their analysis of that indicates that, while the speaker continued to use that deictically (to refer to a visible object), anaphorically (to refer to something the interlocutor just said), and as part of formulaic phrases (e.g., That's the way it goes!), comprehension of later uses was problematic because the referents appeared to be “only in the speaker's mind” (p. 72). Functional analyses indicated that these lexical items were useful on the social level as well as on the more expected informational level (reference); indeed social, rapport-sustaining functions took on heightened importance as communicative abilities on the information level declined. These observations, in addition to those of Hamilton (Reference Hamilton2008), Maclagan and Mason (Reference Maclagan, Mason and Davis2005), and March et al. (Reference March, Wales and Pattison2006) argue that coding of “empty” language as part of a semantic richness analysis may well be too simplistic and needs to be carried out with an understanding of the discursive, social, and physical environments.
Focus on Nonverbal Communication
Turning now to aspects of communication that contain no propositional or verbal content, such as gesture, intonation contours, and laughter, recent work shows fascinating new connections to the role of automaticity discussed in Hamilton (Reference Hamilton1994a), where the more automatic a language feature or task has become over the experiences of a lifetime, the greater the likelihood an individual will be able to (continue to) use it appropriately. Because verbal communication tends to decline over time in dementia, building on automaticity is promising, in terms both of caregiver response and speaker resources for communication and identity construction. This relatively automatic communication can be nonverbal but vocal, such as intonation contours and laughter, or nonverbal and nonvocal, such as gestures and ways of moving the body.
Even when speech is incoherent and void of linguistic meaning, automaticity is displayed in the smooth and appropriate exchanges of turns-at-talk that consist of intonation contours that can (apparently) be recognized, repeated, and/or responded to. Kontos (Reference Kontos, Leibing and Cohen2006) described the following interaction as two individuals with dementia were about to eat breakfast at two different tables in a long-term care facility's dining room.
Abe: [sat down] Bupalupah! [shouting]
Anna: [twisted around in her chair to be able to look at Abe]
Abe: [eyes wider, smiling] Brrrrrrr! [shouting with rising to falling pitch]
Anna: Brrrrrrr! [imitating volume and intonation contour]
Abe: Bah! [shouting, then pausing while looking at Anna]
Anna: Shah!
Abe: Bah!
Anna: Shah! [turned back around in her chair with her back to Abe]
Abe: Bupalupah! [shouting]
Anna: [raised one arm about her head, lowered it in a swift motion; both then began to eat breakfast] (Kontos, Reference Kontos, Leibing and Cohen2006, pp. 206–207)
Kontos (Reference Kontos, Leibing and Cohen2006) claimed that this exchange illustrates Merleau-Ponty's (Reference Merleau-Ponty and Edie1964, p. 7) argument that “communication dwells in corporeality or, more specifically, in the body's capacity to gesture.” Importantly, the speakers continued to exhibit intersubjectivity and displayed aspects of self-identity through their bodily posture and the repetition of syllables, volume, pitch, and intonation contours.
Following studies in aphasiology that have focused attention on the employment of nonverbal behaviors to supplement verbalization (Simmons-Mackie & Damico, Reference Simmons-Mackie and Damico1997) and based on her observations of elderly Swedish speakers in a Finnish day care center, Lindholm (in press) investigated how individuals with dementia used laughter in sequences that appeared to be problematic. By carefully analyzing the placement of laughter within the proverb games already described, as well as the nurses’ responses to this laughter, Lindholm was able to show the importance of laughter in managing face threats linked to language and memory problems. In her study, individuals with dementia used laughter in sequences that appear to be problematic, either in terms of production or comprehension. In some cases, this nonverbal behavior supplemented verbalization in the form of explicit comments on difficulties or initiation of repair; in other cases, the laughter stood alone. Because the nurses did not usually laugh in response (and therefore seemed to understand that the laughter was managing face problems and not signaling joy or humor), Lindholm claimed that laughter was used successfully by these speakers to demonstrate their inability to carry out the task requested of them.
Moving from vocal to nonvocal communication, a number of scholars point to the importance of body language, for the individual with dementia as well as for the healthy conversational partner. In the former case, nonverbal and nonvocal behaviors acquired and used over a lifetime are intimately connected to the display of the individual's identity or sense of self. In the latter case, these behaviors can be important and comforting resources in the relative absence of verbal communication. To illustrate, Ignatieff (Reference Ignatieff1999) wrote the following about his mother: “It [my mother's disease] changed my view of what a person is. . .. It taught me, for example, to be less sentimental about memory as a carrier of human continuity. My mother had no memory whatever, but she was the same person. There were continuities.” Some of these continuities can be discovered by a “heightened awareness of body language” (Chatterji, Reference Chatterji, Leibing and Cohen2006, p. 218) that may include a focus on tone of voice, facial gestures, mime (Snyder Reference Snyder, Hughes, Louw and Sabat2006, p. 260), and even muscle memory (Kontos, Reference Kontos, Leibing and Cohen2006). Importantly, this focus on the nonverbal allows individuals with dementia to communicate emotions such as anger, joy, and empathy in response to interpersonal interactions even when they are unable to communicate propositional content.
In her moving descriptions of individuals in a long-term care facility, Kontos (Reference Kontos, Leibing and Cohen2006) highlighted the fact that muscle memory enables certain actions or gestures by these individuals to “give off” (cf. Goffman, Reference Goffman1959, for the distinction between information “given” and information “given off”) social meaning even in the very latest stages of dementia. Kontos described a woman who was unable to feed herself but who still attempted to take care of her physical appearance:
As a health care aid was feeding her, cereal dribbled from Molly's mouth and coursed down her chin. When the aide tried to give her another spoonful, Molly wrinkled her forehead and gently pushed the woman's hand away. Molly then lifted her bib to her mouth to wipe away the cereal. It was only after her chin was clean that she accepted another spoonful . . . [later] reaching her wavering hand to the back of her neck, she struggled to pull something from underneath her bib. Extending her arm appeared to cause her pain and discomfort, yet she persisted. She eventually revealed a string of pearls that she was wearing and that had been covered by her bib. She allowed the pearls to pass through her long, slender, perfectly manicured fingers, placing the necklace ever so delicately atop her bib. (Kontos, Reference Kontos, Leibing and Cohen2006, pp. 197–198)
Kontos (Reference Kontos, Leibing and Cohen2006, p. 206) maintained that each of these movements was “inextricably movement and embodied selfhood,” that is, that these muscle movements to clean off her chin and to display her pearl necklace worked to construct Molly's self-identity. These movements allowed Molly to highlight a more positive aspect of herself as she attempted to manage her identity in the fact of incontinence, helplessness, and dementia. Observations of this sort are available to attentive and attuned conversational partners and may serve as meaningful evidence of the “continuities” self-described by Ignatieff (Reference Ignatieff1999).
Next Steps: Implications of the State of the Art on Future Research
From the preceding discussion, it should be clear that the intellectual environment is established for highly productive and informative sociolinguistic work at the intersection of language and dementia. Although the need for sociolinguistic work to complement the more prevalent work from neuro- and psycholinguistic perspectives has been known for over 20 years, relatively little work has followed foundational work by Hamilton (Reference Hamilton1994a, Reference Hamilton, Bloom, Obler, DeSanti and Ehlich1994b, Reference Hamilton1996), Ramanathan (Reference Ramanathan1995, Reference Ramanathan1997), and Ramanathan-Abbott (Reference Ramanathan-Abbott1995). Exciting contributions in neighboring fields, including anthropology, sociology, psychology, and philosophy, have set the stage for fine-grained sociolinguistic analyses to relate linguistic details to aspects of the interactional context or to provide insights into emergent discourse.
Identified gaps call out for additional studies of discourse related to dementia within other languages and cultures, with a focus on both public and private spheres, to facilitate an understanding of the role of cultural values and health beliefs in this area. Recordings and transcripts should be made of a greater variety of activities and physical contexts in which individuals with dementia participate to enable fine-grained sociolinguistic investigations. Sociolinguistic studies should be systematically designed with greater attention to variation among individuals so that we can begin to learn more about potential linguistic influences due to sex, first and second languages, professions, education levels, ethnicity, and so on. Because of the linguistic and discursive changes over time that accompany dementia and claims of high levels of day-to-day variability, it is imperative that increased numbers of longitudinal studies be designed and carried out, with an awareness of the heightened role of nonverbal communication to compensate for declining verbal abilities.
Finally, in all this work it is critical that sociolinguistic researchers reach out to the research being carried out within experimental contexts with standardized instruments to bridge the gap between sociological/anthropological perspectives and psycholinguistic/neurolinguistic ones (see also Rabinow, Reference Rabinow2003, p. 18 ff., as cited in Leibing, Reference Leibing, Leibing and Cohen2006, p. 243). Such cross-disciplinary cooperation will open up the possibility of our understanding the range of abilities and disabilities in relationship to individual characteristics, activities, time, and space. This understanding, then, may help in the development and evaluation of treatments and therapies as we work toward solving the myriad mysteries that lie at the intersection of language, dementia, and society.
