Recruitment: Offers, Requests, and the Organization of Assistance in Interaction

ABSTRACT In this article, we examine methods that participants use to resolve troubles in the realization of practical courses of action. The concept of recruitment is developed to encompass the linguistic and embodied ways in which assistance may be sought—requested or solicited—or in which we come to perceive another’s need and offer or volunteer assistance. We argue that these methods are organized as a continuum, from explicit requests, to practices that elicit offers, to anticipations of need. We further identify a class of subsidiary actions that can precede recruitment and that publicly expose troubles and thereby create opportunities for others to assist. Data are in American and British English.

One of the most ubiquitous and abiding features of our social lives, indeed of human sociality itself, is that we need, seek, receive, or are offered assistance by others in small things (e.g., help in opening a tight-fitting lid, someone covering for us when we have to be away from work for a day) as well as large matters (e.g., a significant financial loan). Whatever our circumstances; whatever our domestic arrangements or occupational status; whatever our language and for those without spoken language (or much spoken language; C. Goodwin, 1995); whether in the home, at work, or at leisure; whatever our age or position in a community or society, we rely on the assistance of others to help us accomplish those mundane and not-so-mundane tasks that we would not be able to accomplish by ourselves or which are made the easier for being shared.
Consider, for example, the world of people in later life for whom it becomes necessary to seek assistance with those small things, like making a bed, that they could previously manage by themselves and perhaps that they would prefer to do so still (Lindström, 2005), or when engaged in some collaborative activity such as cooking together, seeing the other crossing the kitchen carrying a pot for the oven, the door of which is closed, one anticipates the difficulty the cook might have opening the oven door and so steps over to open it for him or her. Such minor acts of assistance are essential in enabling us to manage the tasks of our ordinary, daily social lives; without such assistance we might struggle to manage certain tasks at all or to manage them without undue difficulty or inconvenience.
Recruitment lies at the very heart of cooperation and collaboration in our social lives. As a consequence of the reciprocal and collaborative ways in which we can manage to indicate that we need assistance, and the altruism with which we perceive and respond to the needs of others, recruitment is fundamental to the management of social cohesion and solidarity in ordinary social interaction. Quite central to the social sciences is the lineage of explorations into and explanations of how social order and cohesion are maintained and into how cooperation, collaboration, and altruism are possible, given the pursuit of individuals' self-interests. A common theme in this lineage, Heritage (2008) observes, is "that social relations in groups involve a trade-off between competition and cooperation" (p. 310). Our understanding of the mechanisms underlying the balance between competition among individuals and their capacity to cooperate, begins most famously with Hobbes (1651Hobbes ( /1982, for whom human cooperation relies on the trust that each has in the other keeping their side of the "social contract." Adam Smith (1763Smith ( /1978 similarly invoked a bargain between individuals in which "man continually standing in need of the assistance of others, must fall upon some means to procure their help" (p. 347)-a formulation that is especially salient to the sense of recruitment we are proposing here, as is his view that the sympathy we have for others through observing their conduct, their needs, is the moral faculty or mechanism through which self-interest is regulated (Gopnick, 2010).
There are remarkable parallels between Smith's formulation and Goffman's proposal that individuals act in large measure to preserve their "face"; to do so they cooperate to preserve the other's face in order that he or she should cooperate in preserving self's face. Thus considerateness of others (Goffman, 1967, p. 10) emerges from self-interest. Goffman's explication of the role that face work plays in the maintenance of social cohesion, combined for instance with Grice's Co-operative Principle (Grice, 1975), has been central to contemporary theories of cohesion and affiliation in social interaction (Lindström & Sorjonen, 2013), especially the influential theory of politeness (Brown & Levinson, 1987).
Another contemporary theory relevant to explanations for people's considerateness and helpfulness toward others states that "humans have a natural tendency to be helpful and co-operative" and that therefore social relations in social interaction are characterized by altruism (Warneken & Tomasello, 2009a, p. 456). Studied from an evolutionary perspective, much of the research on altruism has been restricted to young children (and chimpanzees) in experimental situations (e.g., Tomasello, 2008;Warneken & Tomasello, 2006, 2009b. Our natural tendency to help is evident from experiments in which even very young children respond spontaneously to difficulties that another encounters by volunteering assistance. In adults altruism has been studied primarily through behavioral games, such as the prisoner's dilemma (Rapaport & Chammah, 1965), as well as social psychological experiments (e.g., Latané & Darley, 1970).

From offering and requesting to recruitment
In one way or another, therefore, most research into altruism in humans has focused either on young children or on adults in economic games using experimental methods. The precise ways people come to help one another in ordinary social interactions have yet to be investigated.
Research in this area has until recently eschewed the fuller spatial and embodied context in which people solicit and receive help, in part because we have tended to focus our research principally around just these vernacular terms: requests and, to a lesser extent, offers (Curl, 2006). This is for two reasons: We have responded, quite reasonably, to speech act theory (Austin, 1962;Searle, 1969); and the act of requesting-a request-has seemed to inhabit a single turn at talk and thereby to be delivered through identifiable linguistic forms. Speech act theory developed as an inquiry into the nature of language, replacing realist theories of language (in which the "meaning" of language derives from its descriptive character and on the truth conditions for describing something with a certain term) with the view that language primarily delivers action. It was therefore concerned with how action is encoded in or delivered through linguistic form; it was not concerned with interaction and how courses of action are managed in interaction. Hence speech act theory was not concerned with the fuller, more holistic spatial and temporal contexts in which such activities as offering and requesting occur, nor with the embodied and material resources through which others are recruited to help. So that though offering and requesting are in many respects social forms, carrying with it implications of need, obligation, reciprocity, imposition, benefaction, and constraint (see Clayman & Heritage, 2014;Couper-Kuhlen, 2014;Drew & Couper-Kuhlen, 2014), offers and requests have been regarded largely as linguistic forms, as being done through characteristic grammatical forms, through certain prosodic resources and suchlike (e.g., Couper-Kuhlen, 2014;Curl, 2006;Curl & Drew, 2008). So speech act theory has been concerned with how to map actions-requests, offers-onto linguistic expressions (see Drew & Couper-Kuhlen, 2014).
The speech act origins of inquiries into an action that has a ready vernacular label, combined with a linguistic approach to offers and requests, has resulted in a somewhat atrophied view of these as singular actions that initiate adjacency pairs (Sacks, 1992;Schegloff, 2007). Indeed, as Curl and Drew (2008, p. 134) noted, research on offers and requests as first pair-parts of adjacency pairs has largely focused on responses to these actions and the preference for acceptance and granting (Atkinson & Drew, 1979;Heritage, 1984;Kendrick & Torreira, 2015;Wootton, 1981). It has recently become clear that in face-to-face interaction offering and requesting do not inhabit just a single utterance or turn, that assistance may be solicited not only through verbal requests or volunteered through verbal offers but also through many other semiotic resources, for the investigation of which it is more appropriate to view the recruitment of assistance in its multimodal and unfolding environment (see, e.g., M. H. Goodwin, 2006;M. H. Goodwin & Cekaite, 2013;Kärkkäinen & Keisanen, 2012;Mondada, 2014a;Rossi, 2014). Moreover, requests have generally been treated as quite independent from offers, as though they inhabit different interactional contexts. We have elsewhere shown that there is instead a symbiotic relationship between requesting and offering, deriving in large measure from specific-and again embodied-circumstances of need and from the possibility that one way of being solicitous, indeed cooperative, is to anticipate the need another might have and either to offer assistance or to give that assistance directly (Kendrick & Drew, 2014). This symbiotic relationship is therefore a further aspect of social solidarity and social cohesion and is a contingent relationship, built upon particular circumstances in which difficulties and thereby needs arise.
In this article we first describe alternative methods for the recruitment of assistance, which we argue form a continuum from the most explicit to the most implicit, and examine the systematic differences between them. We then identify a novel class of actions, which we call subsidiary actions, through which participants attend to and possibly resolve troubles in the realization of practical courses of action and which thereby expose the troubles to public view, furnishing opportunities for coparticipants to give or offer assistance.

Data and collection
The data for this article come from two corpora of video recordings: the Language and Social Interaction Archive (San Francisco State University, n.d.) by Leah Wingard and recordings by Giovanni Rossi made in 2011. Informed consent was obtained from all participants, though the faces of some have been blurred in the figures to protect their identities. All names in transcripts are pseudonyms. The corpora include recordings from a range of social settings, but cases identified for our analysis, including those shown in this article, are all from informal social interactions among friends, family, and colleagues. In all, a collection of approximately 300 cases were identified and were available for and drawn upon in the analysis. The collection centers on the recruitment of assistance as a basic social organizational problem, not on specific linguistic practices or actions. That said, the collection includes requests for and offers of actions and objects, in all forms, as well as embodied and linguistic practices whereby troubles in the realization of practical courses of action become public, be these reports of difficulties or needs, imprecations or exclamations, or visual displays of trouble. Only a small proportion of the cases can be shown here to illustrate the general findings arising from the analysis.

The recruitment continuum
The methods of recruitment that we have identified are organized along a continuum: from those that are largely verbal, which explicitly seek or solicit another's assistance and initiate a course of action to resolve a difficulty (so most like what may be termed requests, albeit accompanied by other bodily and nonvocal actions); to those in which the report or display of a difficulty elicits offers of assistance; and finally to those in which another person's assistance has not been solicited, even indirectly or subtly, but in which the other anticipates a difficulty and acts to preempt it (see Levinson, 2012, andTsuchihashi, 1983, on action continua in other domains). From here, we will designate the one who solicits assistance or to whom assistance is given as Self, and the other one(s) who give that assistance as Other(s). In general terms, the continuum thus spans from the most explicit to the most implicit methods of recruitment. In this section, we examine five distinct methods of recruitment along this continuum-requests, reports, alerts, embodied displays, and projections/anticipations-and consider the systematic differences between them.

Methods of recruitment
Requests for assistance Perhaps the most familiar method for the recruitment of assistance involves an explicit request. If Self encounters a difficulty in the realization of a course of action, he or she can ask Other to perform an action that may resolve the difficulty and thereby allow the course of action to move forward. In the following extract, Graham encounters just such a difficulty: He has rolled a cigarette to smoke but has no light.
When Graham asks "does anybody have a lighter" in line 1, he conveys a difficulty or a trouble that he has in completing a course of action that began with his rolling the cigarette and will end in his smoking the cigarette. The request serves simultaneously to indicate that he means to light and hence smoke the cigarette but does not have the means to light it, as well as to solicit assistance from one of the others with whom he's sitting. Hence the difficulty arises from his not having the means to fulfill a necessary step in, and thereby complete, the projected course of action, to smoke a cigarette. To say that he "needs" a light, as one might do, is therefore not a reference to a psychological disposition (Childs, 2012); it denotes rather a difficulty in the realization of a practical course of action.
The practices that Graham uses to solicit a solution to his difficulty are recurrent and accountable practices for requesting and are among the most frequent such practices in our data (Kendrick, 2015). In each case, the request initiates a sequence in which the provision of an object becomes a specially relevant next action. The first request ("does anybody have a lighter") does not address the request to a particular participant, nor does it assume that the object is available, as other forms do, but it is nonetheless recognizable and accountable as a practice for requesting an object (Fox, 2015;Rossi, 2015). The second request ("can you pass me those matches") is addressed to a particular participant and refers to a particular object, both through the definite noun phrase "those matches" and through a pointing gesture that accompanies it (see Figure 1). The form of the request, a modal interrogative with can, displays little orientation to contingencies that may affect the grantability of the request (Curl & Drew, 2008), as indeed the matches are on the grass directly in front of Daniel and can be passed with ease.
Reports of needs, difficulties, or troubles Through his requests, Graham formulates possible solutions to a difficulty, which he asks those around him to provide. The requests, however, leave the difficulty itself implicit in that they do not report the problem per se (cf. "I need to light my cigarette" or "I don't have a lighter"). Reports of needs, difficulties, or troubles constitute a practice for recruitment in their own right. In contrast to requests, such reports do not establish a normative obligation for the other to assist but rather create an opportunity for the other to volunteer assistance (Kendrick & Drew, 2014; see also Drew, 1984;Vinkhuysen and Szymanski, 2005;Curl, 2006;Zinken & Ogiermann, 2011). 1 In the following extract, as three housemates cook dinner together, a complaint by one provides an occasion for assistance by another.
As Kerry prepares a plate of food on the kitchen counter, she turns to James and reports a minor difficulty, namely that she does not like to eat dinner while on camera (as she had also done the day before). After James does not respond, Kerry self-repairs her turn, replacing "eat dinner" with "feed myself" (line 4). The repair not only upgrades the complaint, in that it casts her actions in an animalistic and unflattering light, it also covertly pursues a response (Bolden, There are at least two classes of need and necessity statements: those that formulate a solution to a trouble (e.g., "I need that pen back actually") and those that formulate a trouble or difficulty (e.g., "I need to plan my Shakespeare lesson") for which the recipient may generate a solution. The method of recruitment described in this section concerns the latter. , & Wilkinson, 2012). In overlap with the repair, James offers to turn the camera off, an action that would resolve the difficulty. He then lowers his hands from the cutting board on the counter and wipes them on his jeans, in preparation to handle the equipment, and then begins to walk across the kitchen toward the camera (see Figure 2). In effect, Kerry's complaint recruits James to assist her in that it creates an opportunity for him to perform an action that would resolve the difficulty. Whereas requests formulate a possible solution to a difficulty and create a normative obligation for the recipient to provide the solution in response (e.g., "put off the camera," "could you put off the camera?"), reports of difficulties, such as Kerry's complaint, do not specify a solution and do not mandate that the recipient should provide a specific solution in response.

Trouble alerts
A method for the recruitment of assistance that formulates neither a solution to a difficulty nor the difficulty itself involves the production of a trouble alert. A set of linguistic practices exists, the production of which alerts Other that Self has encountered a difficulty. These practices, which Goffman (1978) named response cries, include a variety of interjections (e.g., "oh," "oops") and imprecations (e.g., "shit," "oh. = damn it"), as well as other affect-laden sound objects, like shrieks and cries. Trouble alerts signal that there is a difficulty but not what the difficulty consists of. In the following extract, after Kevin signs the researcher's consent form, he hands the pen to Travis so that he can do the same. Travis takes the pen and places the form on his knee, in preparation to sign it.
As he produces "uhm" (which marks a transition from a joke to the task at hand), he holds the paper on his knee and directs his gaze around the area before him for approximately 0.7 seconds. He then returns his gaze to the form, leans forward, and begins to sign it on his knee. It is at this moment that he marks the beginning of the action with "right" and then alerts the others of its  difficulty with "oh god" (line 3). This immediately draws the attention of Kevin, who turns his head and directs his gaze to the form as he leans to his left toward Travis (line 5; see Figure 3). He has his gaze on the form for approximately 0.6 seconds before he holds out a stack of consent forms in a plastic sleeve, which he had used as a surface to sign on, and offers it to Travis verbally ("you want that"). Travis continues to sign the form, not responding to the offer. Kevin then lowers the stack of forms, as if to set them down, but then removes them just before Travis refuses the offer (line 15). As we can see, Travis's trouble alert effectively recruits Kevin to assist him, both in the form of a verbal offer and in the provision of an object that might relieve his difficulty. According to Goffman (1978), response cries "[do] not officially establish a slot which [the recipient] is under some obligation to fill" (pp. 798-799). Although trouble alerts do routinely and perhaps normatively solicit the attention of Other(s), they do not establish a normative obligation on Other(s) to provide assistance, as requests do. Nor do trouble alerts formulate the nature of the trouble; the other must attend to the source of the alert and diagnose the trouble on his or her own. Like reports of troubles or difficulties, trouble alerts create an opportunity for Other to volunteer assistance, as a sequence-initiating action.

Embodied displays of trouble
In comparison to requests for assistance and reports of troubles, trouble alerts are a less explicit method for the recruitment of assistance. Less explicit still are embodied actions that merely display a trouble visually. Such actions can recruit others to assist even when they are not, in the first instance, accountable as forms of solicitation or alerts of trouble. The following case, which comes from an interaction among a group of students in a common area of a university building, demonstrates this point. Here Mark can be seen to encounter some difficulty as he looks across the table at a picture in a book held by Rachael, who then holds the book up for him to see more clearly.  An assessment of the picture begins the sequence (line 1) and draws the attention of Connor, seated to Rachael's right (Figure 4a). In turn Connor's surprise (line 5) draws Mark's attention; Mark then leans forward and gazes at the picture from across the table (line 7). Mark holds this position for approximately 0.5 seconds and then torques his head back and to the side, such that the orientation of his head comes to approximate the orientation of the book (see Figure 4b). The torque of Mark's head makes publically available a minor trouble, namely that from his perspective, seated on the other side of the table, the picture appears upside-down and would therefore be difficult to see. Mark's head movement attracts Rachael's gaze (line 11), at which point she would be able to see his head in an unstable position and his gaze directed to the picture. Shortly thereafter she lifts the book and holds it up for Mark to see (line 14; see Figure 4c) and thereby resolves his trouble. Mark's visible bodily actions expose the trouble, making it publicly available, and thereby provide an occasion for Rachael to assist him, voluntarily. The actions in effect recruit her, even though in the first instance they are recognizable and accountable as an action taken by Mark to resolve the trouble without assistance.

Projectable troubles
At the far end of the continuum are recruitments in which Self produces no action, whether linguistic or embodied, to solicit assistance, nor does Self report, alert, or display a trouble. Nonetheless, Other anticipates a trouble in a course of action by self-before it manifests-and acts to preempt it. Such cases of anticipatory assistance exploit the projectability of courses of action, as in the following extract, which comes from an interaction between friends as they play a board game. After a phone rings in an adjacent room, Becky announces that it is hers and stands up from the table to answer it (see Figure 5). As she begins to walk forward toward Shannon, whose chair blocks her path, Shannon slides her chair forward to let Becky pass.  The ringing of the phone and the announcement by Becky that it is hers project in advance a particular course of action, namely one in which Becky will walk around the table and go into the adjacent room to answer the phone. As Becky begins to realize this course of action, standing up and taking a step toward the corner of the table, Shannon evidently comes to recognize an imminent trouble: Her chair is in Becky's way. Shannon reaches down to grab the sides of her chair, leans forward, and begins to slide her chair forward. As she does so, Becky apologizes, presumably for the inconvenience and the interruption to the game, and Shannon responds with an apology of her own (lines 12-13). Crucially, the apology by Becky comes after Shannon has recognizably begun to slide her chair forward (see line 11, which shows that the preparation of the movement begins approximately 0.5 seconds before the apology). The apology is therefore not a solicitation of assistance but a response to it. In this case, there is no request, no report of a trouble, no alert, and no display of difficulty. Shannon anticipates a trouble, one that necessarily involves her, in the course of action initiated by Becky and acts preemptively to obviate it.

Comparison of methods
As the extracts in the previous section demonstrate, if Self encounters a trouble in the realization of a course of action, he or she can select from a set of alternative methods, a possible outcome of which is a resolution to the trouble in the form of assistance by Other. We now describe the differences between the methods in more general terms.
Each method on the recruitment continuum differs in how the trouble becomes recognizable. Requests do not formulate or display the troubles for which they solicit resolutions (though the accounts that occasionally accompany them frequently do just that: see Kendrick, 2015). That Self has a trouble or need is conveyed by the request, in that it is recognizable as a solicitation of assistance, but the nature of the trouble is left implicit. Trouble alerts, too, index a trouble but do not formulate it explicitly. Such alerts signal that a difficulty or need has arisen and that Other should attend to the source of the alert to identify the nature of the trouble. In contrast to these methods, reports by definition formulate troubles explicitly in a turn at talk. The visible bodily actions that Self performs to realize a course of action can expose difficulties to public view and thus come to serve as displays. In such cases, troubles are not so much indexed or formulated but embodied. Finally, projectable courses of action can allow Other to anticipate a need and act to preempt it before it occurs. The trouble has no form per se, whether linguistic or embodied, nor does Self's actions signify the trouble.
The methods on the recruitment continuum also differ in the relevance of assistance as a response. A distinction can be made between requests for assistance on the one hand, which initiate adjacency pairs in which giving assistance is a relevant response, and on the other hand all other methods. Whereas Self's request for assistance establishes a normative obligation for Other to produce a specific assisting action, reports of trouble and trouble alerts create an opportunity for Other to give or offer assistance. That this is so can be seen in the forms that Other uses in response to such reports and alerts. Note that "should we put it off now" (Extract 2) and "you want that" (Extract 3) are designed as sequence-initiating actions, namely offers, not as responding actions. The Other does not comply with, grant, or fulfill a request for assistance; she gives or offers assistance agentively and voluntarily.
The distinction between requests and other methods for recruitment also concerns who initiates the recruitment of assistance, not only in terms of sequence organization but also in terms of who presents a possible resolution. Requests are explicit formulations of actions or objects that Other should perform or provide to resolve a trouble or satisfy a need. In other words, Self presents a possible resolution to Other (e.g., "can you pass me those matches"), who should then implement it (e.g., by passing the matches). The other methods for recruitment, in contrast, have a different organization: Actions by Self enable Other to recognize or anticipate a trouble, but these actions do not present Other with a possible resolution; Other must generate this independently. This is one sense in which the assistance is voluntary. That said, the particulars of each case can constrain, to a greater or lesser degree, the set of possible resolutions. A complaint about the unwelcome effects of the researcher's camera affords particular solutions but not others, and the difficulty of signing a form on the grass or one's knee affords fewer solutions still. But insofar as the Other does provide or offer a resolution, requests aside, it will have been generated by Other. Figure 6 presents the recruitment continuum in schematic form.
Finally, the solutions that Other provides manifest as offers of assistance (Extracts 2 and 3) or direct provisions of assistance (Extracts 4 and 5). In this way, the recruitment continuum places requesting and offering into a systematic relation, as alternative actions initiated by Self or Other for the recruitment of assistance and the resolution of troubles in the realization of practical courses of action. Recruitment therefore concerns not only the methods by which Self solicits assistance from Other, no matter how subtle or covert, but also the methods by which Other comes to recognize Self's trouble, difficulty, or need and acts to resolve it.

Subsidiary actions preceding recruitments
Generally speaking, requests for assistance are not initial actions. There may be a sequence or a series of actions, verbal as well as embodied, that precedes and results in a request being made (see, e.g., . If recruitments of assistance are situated more holistically, in the context of an unfolding scene and sequence of embodied actions, they have antecedents. It is evident in our collection that recruitments are commonly preceded by visible bodily actions by Self that in one way or another attend to and expose a difficulty, trouble, or need and thereby create an opportunity for Other to offer or provide assistance. We refer to such actions as subsidiary actions because they support or facilitate the resolution of a difficulty and hence the realization of the primary course of action that has been impeded. In this section, we illustrate three alternative trajectories of action, each beginning with a subsidiary action by Self. We show that subsidiary actions can (a) result in an independent resolution of a trouble by Self, (b) precede a solicitation of assistance by Self, and (c) precede and occasion a provision of assistance by Other.
Subsidiary actions are, in the first instance, actions by Self to attend to and possibly resolve troubles, needs, or difficulties independently, without assistance by Other. In the following extract, a participant who encounters a problem in the realization of a course of action produces a gestalt of visible bodily actions as she searches for a resolution. As a group of friends prepares a barbeque, Alison halts or "freezes" and, placing the tips of the fingers of her left hand on her chin and furrowing her brow, adopts what is quite transparently a "thinking face" (M. H. Goodwin & Goodwin, 1986). She holds this, virtually motionless, for approximately 1.4 seconds. Each of the others is turned away from Alison as she turns her head slowly to the right, a movement that takes approximately 0.6 seconds.  Figure 6. The recruitment continuum. The methods for recruitment differ in how the trouble becomes recognizable, the relevance of assistance as a response, and whether Self or Other initiates the recruitment.
Alison's visible bodily actions can be seen as a display of puzzlement, though the cause or object of her puzzlement cannot yet be determined. However, as she turns her head, we can come to see her actions as a visual search of the environment-she's looking for something. While this is evident to us, as analysts, from the recording, it could not have been evident to the others because they are each engaged in other matters and turned away from Alison-so that her "visible" display of trouble is not visible to them. The visible bodily resources that she uses to conduct the search, her facial and manual gestures, her head movement, allow it to be recognizable as such-she's not only looking for something; she's doing looking for something. After a search of approximately 2.0 seconds, she apparently spots the sought-after object, a lighter, and announces the end of her search with "here it is" (line 11). This announcement, like her gestures, orients to the accountability of her actions (Garfinkel, 1967), even though, as we can see in Figure 7, only the eye of the camera is on her as she conducts her search. She then reaches out and picks up the lighter, walks around to the front of the table, and sets it down as she marks the completion of the course of action with "okay." The lighter is later used to light the coals in the grill.
Here, then, we see that a difficulty has arisen in the initiation of a course of action, Alison's project involving the lighter. After a visible display of that difficulty-a subsidiary action through which she searches for an independent solution-she resolves it on her own; her visible bodily actions did not recruit others to help, perhaps because they were unable to see them. In the following extract, the conduct of Self that denotes a difficulty is quite visible to Other, but what difficulty the  +,,,,,,,,,,, conduct denotes is opaque. Two friends are sitting outside on the grass enjoying the sunshine when Liz begins rubbing her upper lip with the forefinger of her right hand. But this is not unambiguously a sign of discomfort, as she first touches her lip and holds it momentarily as though a display of thinking similar to the previous case (for a similar account of alternative possibilities of "thinking" or "requesting," see C. Goodwin, 1987, p. 118).
Up to this point, while Liz's conduct is visible to Charlotte but possibly ambiguous about denoting a difficulty, Charlotte is looking directly at Liz. But as Liz makes small rubbing movements back and forth across her lip, which might help to disambiguate her gesture, Charlotte turns away. Hence to begin with, Liz's visible bodily action is opaque as regards a difficulty; at the point when she adjusts that action to (a little) more clearly indicate a difficulty (discomfort), Charlotte turns away from her, so that Other has either not noticed Self's need or has not recognized the difficulty and need denoted by the conduct. That difficulty/need becomes evident when, Other having missed an opportunity to offer assistance, Self asks for the moisturizer (line 28). Note that in doing so Liz asks to borrow "a bit more of" Alice's moisturizer, suggesting that she has borrowed or used it previously, and that her visible actions might have been understandable in light of that previous use.  Indeed, even as Liz's request is in progress, Charlotte already directs her gaze, observed in her head direction, toward her bag before she reaches out to retrieve the moisturizer (line 29). However, the subsidiary actions by Self have, once again, not resulted in a recruitment of Other to help, so it falls to Self to initiate the recruitment through an explicit request. These cases are representative of many in our collection, in which a visible bodily action by Self is associated with and signals a difficulty that will impede a course of action. That difficulty is displayed by Self's visible conduct, although not always in a way that is registered or understood by Other(s), either because that conduct is ambiguous or opaque as to whether it denotes a difficulty or because it is not visible to Other(s). So that while Self's bodily actions are trouble attentive in a rather visible way, such actions may or may not recruit Other to assist.
Although one cannot know what intentions might lie behind such actions, it is evident that they create opportunities for Other to give or offer assistance. If attended by Other, subsidiary actions that Self performs to resolve a trouble are recognizable as such and therefore expose the trouble to public view. For a visible bodily action by Self to elicit help, Other must visually monitor Self's actions (see Extract 4). As displays of trouble, visible bodily actions are therefore dependent on Other's line of sight and deployment of visual attention. But such displays are not limited to the visual modality, as the following extract shows. Here, while a group of people have drinks before they go out to a street fair, Dustin picks up his glass, which contains ice cubes and the last remains of a drink, and gently twirls it repeatedly, causing the ice cubes to rattle. Tyler, who is looking down at his phone while this occurs, evidently hears the rattling, quickly looks up from his phone to Dustin, and extends his hand out toward the glass as he stands up and leans forward. Dustin hands the glass to Tyler, who takes it, walks into the kitchen, and fixes him another drink.  In effect, the rattling of the ice cubes recruits Tyler to act on Dustin's behalf. The sound of ice cubes rattling is a sign of an empty glass, which in the context of the activity in which the participants are engaged-a "pre-party"-is itself a sign of need. This is evidently recognizable to Tyler, who orients to the sound as an occasion for assistance. But there is evidence that, for Dustin, the action he performs is in the first instance an instrumental one, designed not to fish for a refill but to settle the remaining liquid before he takes a final sip. After he twirls the glass, he begins to raise it, in the direction of his mouth, as if to drink (see Figure 9), but he abandons this when Tyler reaches for the glass. The action Dustin performs can thus be regarded as a subsidiary one, through which he prepares to consume the remains of his drink. In this case, the subsidiary action gives off an audible sign that allows Other to recognize Self's need without attention to his visible bodily actions, through the auditory modality alone.

Discussion
We have proposed that "recruitment" provides a more holistic way of conceptualizing and studying how help is sought and volunteered in ordinary face-to-face interaction. Recruitment encompasses both the linguistic and embodied semiotic resources through which Others are recruited to help resolve difficulties. It also encompasses our anticipating another's need of assistance and offering or simply giving that assistance-Other is thereby recruited to assist, without their assistance having been sought. We have emphasized that "need" is not a psychological disposition but refers rather to the difficulty or impedance that can arise during a course of action. Self may encounter a difficulty in completing a course of embodied action, when for instance she cannot find a lighter with which to light the barbeque, when trying to apply a moisturizer to soften dried lips, or trying to sign a form. Hence "needs" arise from the steps that are necessary in order to fulfill a given course of action. We have shown that when encountering difficulties in progressing a course of action, Self may perform certain subsidiary actions to (attempt to) resolve the difficulty, such torqueing the head to get a better view, visibly searching the environment, or twirling a glass of ice cubes in preparation to drink. These visible bodily actions (and their audible signs) display the difficulties Self is experiencing, exposing the trouble in such a way as to recruit assistance by Other(s). Alternatively, by projecting what might occur during an unfolding course of action, Other may anticipate a difficulty Self is likely to encounter and volunteer help either by offering or by directly giving it. We have represented the range of ways in which Other may be recruited, from the ways in which Self most explicitly asks for or solicits help to the unsolicited help given by Other, as a continuum. The recruitment continuum consists of the explicit and implicit, verbal and embodied, Self-or Other-initiated conduct through which recruitment of assistance is effected.
The question of who benefits from offers and requests has been raised by Couper-Kuhlen (2014) and Clayman and Heritage (2014). The obvious answer for recruitment, it would seem, is that Self benefits from Other's assistance. But given that social interaction is collaborative in nature, with activities that involve multiple participants, an action that benefits Self may benefit Other(s) in turn. The search for a lighter in Extract 6 may well be the project of an individual, but insofar as the lighter comes to be used to light the coals for a communal barbecue, the resolution of the search benefits all participants. The difficulty in signing a form in Extract 3 is arguably an individual one, but because it belongs to the preparatory phase of an activity (recording a conversation for scientific research), its resolution facilitates the realization of the activity as a whole and thereby benefits the others as well. In social interaction the question of who benefits from an action is not easy to answer, nor is the relevance of the question always apparent because the actions of individuals constitute activities constructed by multiple participants. It is thus not clear to us on what basis and to what extent the benefits of assistance might be said to belong to Self as an individual or to Self and Other(s) as a collective. Even where benefit could be assigned, such as the request for a lighter to light a cigarette in Extract 1, the question of who benefits does not shed light on how Self and Other manage the recruitment, nor does it shape or organize the alternative methods for recruitment, which arise directly from perceptible troubles, difficulties, or needs.
Our analysis of recruitment suggests a social organization of assistance in interaction in which the practices employed by Self and Other cooperate systematically to resolve troubles in the realization of courses of action (see Schegloff, Jefferson, & Sacks, 1977 for this Self/Other distinction in another organization for the resolution of troubles). Although we have only begun to reveal and describe the practices and principles that constitute this organization, some aspects of its operation, which we will now enumerate, have already become apparent. (a) Self need not know how to resolve his or her difficulty to arrive at a solution. To make a request requires that Self generate a possible solution to a trouble in the form of an action for Other to perform, but the difficulties one encounters do not always have ready solutions.
The organization of assistance includes practices that Self can employ to indicate a trouble and to allow Other to provide a solution. (b) As an extension of this, there is a motivation for Self to report, alert, and display his or her troubles to Other insofar as such actions can lead to trouble resolution. (c) The observation that recruitments follow subsidiary actions suggests that the organization of assistance is such that the independent resolution of a difficulty can take precedence over recruitment. In other words, asking for help can be a second alternative to resolving a trouble on one's own. (d) Because subsidiary actions furnish opportunities for Other to assist, a motivation exists for the ritualization of subsidiary actions into communicative signals.
A visual search of the environment to locate a necessary object may be performed so as to increase its recognizability as a subsidiary action and thereby enhance the opportunity for assistance (e.g., Extract 6). So too may instrumental movements of the body be exaggerated or repeated (e.g., Extract 8). (e) As a final point, the timing of assistance is variable. In some cases, Other anticipates a trouble before it occurs (e.g., Extract 5), while in other cases, assistance appears to be withheld, even as Other monitors the trouble and its management by Self (e.g., the 0.6 seconds in which Kevin watches Travis sign the form before he offers assistance in Extract 3). This implies that the organization of assistance includes principles, which await description, that constrain the assistance of Other. The observations enumerated here, though brief, suggest that the practices employed by Self and Other cooperate in a social organization for the management and resolution of troubles, difficulties, and needs that arise in practical courses of action.
The sheer ubiquity of recruitment in our daily social lives cannot be overstated. It is hardly possible to be copresent with others, to share a space with them, to interact with them, in whatever kind of setting-whether domestic, informal, and social; in the workplace; in an institutional setting such as a welfare office, a school classroom, a business meeting-without needing at some point to enlist someone's help or for someone to help, unsolicited, when they think we might have a difficulty.
Moving one's chair out of the way of someone passing, seeing that another's writing instrument is not working and offering a replacement, having difficulty opening a door or taking off a jar lid, being unable temporarily to find what one is looking for, needing to rest on something flat a document that we are trying to sign-we encounter all such difficulties or impedances with very great frequency in our everyday social lives. All such difficulties and their attendant solutions require small adjustments to be made, both by the one requiring help and the one providing it, adjustments that can involve the (usually small) sacrifices that being recruited entails for the one giving the help -hence the altruism that inhabits recruitment.
Recruitment, whether achieved through soliciting another's assistance or through the voluntary and altruistic help given by others, lies at the heart of cooperation in social life. Indeed, the methods along the recruitment continuum, the symbiosis between offers and requests, and the recognizability and consequentiality of subsidiary actions all constitute specifications of the very notion of cooperation in social life. It is hard to imagine what our daily social lives would be like if we were not able to enlist others' help in resolving such minor matters as those mentioned and illustrated in our data extracts. Hence the cooperative and altruistic character of recruitment can be said to underpin the maintenance of social cohesion at the micro level of social life.