Achieving Preallocation: Turn Transition Practices in Board Games

ABSTRACT This paper contributes an analysis of practices for managing a preallocated turn-taking system in board games, expanding existing studies of preallocation beyond question-answer sequences. Although board games have existed for thousands of years across human cultures, and despite being a widely used method of data elicitation in many fields of research, there are few studies of how adults accomplish play. Using conversation analysis, this paper demonstrates how participants organize transition between board-game turns, finding that participants treat the game turns as analogous to the organization of pre- and post-possible completion. However, the preallocated nature of game turns results in alternate sense making concerning delays and overlap, especially where such occurrences threaten the achievement of the activity. Data are in English.


Methods
Data are in several dialects of English. The data in this study are from three different game play settings: games collected at gatherings of board game hobbyists in homes or cafes, games posted online as "live plays" or nonscripted example plays of the game, and games by novices in university classrooms. Local participants gave informed consent to be recorded. Online data are treated as publicly accessible television given that view counts for any given video are between ten thousand and several million. This study was approved by the ethical review board at Loughborough University. A summary of the data can be found in Table 1.
The data comprise observations from 21 different games, many of which may be unfamiliar to the reader as most are what is sometimes termed "new wave tabletop games" (Jolin, 2016). Although these less-familiar games often have different mechanics than "mainstream" board games (Scrabble, Monopoly) or card games (poker, bridge), they involve the same turn-taking requirements (with the exception of games in which everyone plays at once, which are set aside for future research). Using "new wave" games allowed for more diversity in the game mechanics and the turn procedures analyzed. There is not space to elaborate every game's rules; however, links to information on the games can be found in the Notes. Rules relevant to the sequence at hand will be explained. The findings in this paper can only be taken to apply to turn-based games wherein a one-at-a-time rule is achieved (rather than speed games wherein players all act simultaneously, as if racing).
As the goal of the study is to examine how participants manage preallocation for themselves, this study uses conversation analysis (inclusive of multimodal practices), which is particularly appropriate for demonstrating how participants orient to and treat practices and actions during interaction (for transcription, see Mondada, 2018; for conversation analytic methods, see Sidnell, 2010). Each game turn in the recordings was examined to see which practices participants used for achieving a game turn. In other words, using the ethnomethodological framework that conversation analysis employs, the analytic process questioned how any given practice was treated as a game turn. Deviant cases were included, because they demonstrated what occurs when participants do not adhere to the norms of transition and in which ways such breaches were made accountable. For example, the analysis below includes instances wherein participants treated a game turn as complete when it turned out not to be (Extract 1) or wherein participants skipped part or all of another's game turn (Extracts 15,18).
Novice players are represented in the data (some are whole groups of novices, although only in the university student data, whereas, some sessions in the rest of the data are novices mixed with hobbyists). The online data are a produced media product, as with televised interview data; however, in being nonscripted the players must still rely on their everyday sense-making procedures for management of turn-taking. None of the online data were specifically how-to videos and some were from live-streamed game play.

Analysis
This paper will cover how players managed transition between turns once they came to a recognizable completion point. Following Schegloff (1996, p. 97), 1 the transition space is the focus, and the turns are treated as the time between "episodes of determination of when the action shall pass to another." What constitutes a recognizably complete game turn is beyond the scope of this article, although it will be touched upon. I will first demonstrate that players treat game turns as having recognizable end points and that they can project transitions based on ongoing contingencies. Game turns are the analogous unit for a turn-at-talk; however, games have preallocated means for determining what counts as a complete game turn-that is, for determining when a point of possible completion has been reached. This is in contrast to the Sacks et al. (1974) systematics, in that games do (often) fix turn size and specify in advance (much of) what players "say" (Sacks et al.,Items 6 and 8). What players may "say" constitutes the set of possible game "actions" they may take, such as playing a card, moving an army a certain number of spaces, or drawing new units or cards. Describing the means for preallocating turn size and content is beyond the scope of this paper, but in its simplest terms, games require that certain game actions must happen on a game turn, certain additional game actions may happen. Some games are structured so players do all of a specified set of actions (e.g., draw, move, attack), others allow for, for example, three actions of any combination (three draws, or two moves and attack, etc.). These game actions and constraints on their use provide for the projectability of game-turn completion, much as syntax, prosody, and pragmatics do for TCUs in everyday talk (Ford & Thompson, 1996). The order of those game actions within the game turn is typically preallocated as well; however, the length of real time taken for these events is variable. Let us begin by demonstrating that (a) game turns have a projectable possible completion and (b) next players orient to avoiding starting the next game turn in overlap.
Extract 2: DT Small World SI_14.40  Points of possible game-turn completion are recognizable by players and oriented to as turn relevance places. However, the temporality of such game actions is highly projectable-that is, once the action of placing a token is initiated, it should be evident that the game turn is about to end (Mondada, 2006), much as syntax, for example, can project turn ending by virtue of clause structure (Auer, 2009;Ford & Thompson, 1996). This raises the question of why players do not visibly project further in advance. Across the corpus, players do not typically prebegin the next game turn while the current game turn is still "undecided" (that is, choices such as token location, which card to use, or so on are not accountably complete). I suggest this is due to two factors. First, to prebegin a next game turn early reveals the next player's game choices, which may be strategically disadvantageous. If the current player has not yet accountably completed their game turn, they could alter their move based on visible early prebeginning of the next player's game turn.
The second factor is structural: The completion of game actions is actually only a pre-possible completion (Schegloff, 1996, p. 83), not the possible completion point. The accomplishment of the relevant game actions is the point at which the game-turn ending is projectable. It is the ensuing practices of withdrawing from the play space that accomplish the possible completion point (or as close to a "point," a single moment in time, as one gets in interaction). I suggest this is the case given that the next player is not accountable to undertake the next game turn until after the practices of withdrawing from the play space are done. Such accountability is available in Extract 3. Here, next players overlap their game-turn initiations with the withdrawing practices of the current players, until Line 14. Mammal does not overlap Insect's game turn with any indication that he or she is about to begin the next game turn. However, the moment Insect is fully withdrawn from the play space (L13-14), Mammal begins practices to account for the lack of play (L15-20), showing Mammal's orientation to the withdrawal as the moment at which turn-taking efforts are accountably due.
Extract 3: 160908 Dominant Species 2_1:37  Although Insect does not overlap with an extension of a hand, Mammal and Insect do overlap with an orientation to the board, leaning in and displaying attention to the proceedings and incipient action (see also Extract 10, for a failure to pay attention, followed by a lean in). Mammal does not display incipient action in overlap; however, both players do display incipient action. By hovering their hand with the game token (L16, 18), as well as uttering "uh" (L17), Mammal displays attention to the game, awareness that their game turn is incipient, and projects that play is imminent. In other words, Mammal orients to their accountability for not playing by displaying incipient play. Play is therefore accountably due when the prior player has fully withdrawn from the playing space (L14), not just when the prior player has accomplished the last game action on his or her game turn (pre-possible completion). Withdrawal is not only a matter of reaching (Sacks & Schegloff, 2002) but of making oneself unavailable to play, thus passing on any further activity (in contrast to Extract 1, where withdrawal was not completed). Once the current player has withdrawn and has both reached "home position" and displayed unavailability, failure from the next player to display incipient play results in problems in intersubjectivity and in being sanctioned by other players.
To summarize, the analysis in this paper claims that game turns have an analogous structure to everyday turn-taking, particularly with respect to Schegloff's (1996) turn-transition framework. Players precomplete their game turn by accomplishing final actions (as specified by the game) of their game turn and reach a possible completion point by withdrawing from the play space. The next player is accountably due to play when the withdrawing practices are done, although next players may prebegin, "in overlap," once the prior player has reached pre-possible completion. In the next sections, I will elaborate on the evidence for this structure of game turns. I will first examine the practices players use for withdrawing from the play space and reaching the completion point and, subsequently, analyze the practices players use to account for, or to make others accountable for, delays in taking the next game turn on time. Finally, I will analyze (deviant) cases of interjacent overlap (Jefferson, 1986), in which players prebegin too early and are sanctioned.

Practices for ending a turn
Once the game turn has reached pre-possible completion, players confirm they are completing a game turn by withdrawing from the playing space and making themselves inaccessible for play. The game turn may be visibly complete through evidence on the board, and the next player's (pre)selection likewise often made visible through board-game pieces (see also Lynch et al., 1983, on artifacts as mnemonics for activity progress). Nevertheless, players must still display their orientation to the progressing game turns and are accountable for upholding turn transition. By removing all signs of incipient play, players display that they cannot take further action, repair, or restart. For example, in the next extract, once Steve completes his game actions by placing two cubes (L7), he then enacts several practices to reach a possible turn completion (L8-10), to which the next player, Gordon, responds by initiating the next game turn.
Extract 4: BGZ Lords of Waterdeep_19.12 Only the last of the three practices that complete Steve's game turn are treated as a definite completion point. When Steve leans on the table (L10), Gordon extends and then hovers his hand with a token (L11), initiating the next game turn and treating the leaning on the table as the completion point of the game turn. The (seemingly self-confirmatory) "Yep" (L8) and picking up cards (L10) provide a sequential context of pre-possible that project Steve's game-turn completion point. These practices serially display Steve's orientation to progressivity (Stivers & Robinson, 2006). When Steve finally leans on the table (L10), any doubt of his availability to play is removed and Gordon ratifies Steve's display as one of completion by uttering an activity-shift marker (Beach, 1993) (L12) and starting the next turn.
Leaning back and away from the playing space (the table and game board) is similarly effective, as it mirrors the embodied withdrawals from turns in everyday talk (Li, 2014;Oloff, 2013;Reed, 2015). Withdrawing one's hand from a token was already a visible practice in Extracts 1 and 2. Leaning further away once the hand is withdrawn is a common follow-up, as seen below. Here, Jess passes on the (optional) opportunity to purchase an available property in Monopoly and so completes his game turn without the typical purchasing game action. Jess accomplishes this by both leaning back from the table (L8) and by a "tossing" gesture (L8), in which he gestures as if passing the game turn metaphorically to Steve.
Extract 5: G Monopoly_15:20 ((Steve is re-enacting a conversation in his utterances L1-7)) Jess faced a problem in reaching a completion point; he could not perform game actions that served as pre-possible completion, because he had reached the point whereby all that was left to do (if he was not buying a property) was nothing. The absence of action (or game action) may have displayed accountable delay rather than possible completion. As a result, Jess needed to indicate his choice to not act. The toss gesture and removing himself from the playing space together were one way to accomplish this. Although Steve initiates repair on Jess's withdrawal (L9), he projects that Jess is completing the current game turn by prebeginning his own next game turn by picking up the dice (L10). Steve withholds rolling the dice until Jess confirms the game turn to be completed. As another example of a player leaning away to indicate withdrawal, in the following, Bird not only withdraws their hand but also then leans their full body away from the playing space and sinks down into their chair. Although Reptile is not the next player, Reptile still orients to Bird as having completed his game turn: Extract 6: Dominant Species 1_22.58 Reptile orients to Bird's game-turn completion twice, first, by verbally accepting the game action with "Kay" (L5; note that this may also be an activity shift marker, Beach, 1993) and, second, by returning the token Bird was using to Bird (L8; this is a different stage of the game than Extracts 1 and 2, and now some tokens are retrieved instead of placed). Reptile cannot play, as they are not the next player, so the only way to orient to Bird's game turn being completed is to assist in its completion and/ or to comment on the game turn. It is not until Bird leans away from the playing space, however, that Reptile passes Bird's token over. To move the token during another player's game turn is highly inappropriate, as it would constitute making decisions on the other player's behalf (unless requested). Once the game turn is at a completion point, however, then it becomes possible for other players to collaborate to help to demonstrably close the game turn. If the piece remained on the table, Bird's game turn might still appear "live." To remove the token, especially as it is somewhat out of Bird's reach, is to be supportive and collaborative. However, this is only possible, and only done, when  a player has fully withdrawn. Bird's withdrawal, while demonstrating game-turn completion, ironically also makes the token all the more inaccessible to remove. Besides removing oneself from accessibility to play, players can also verbally announce the completion of a game turn. For instance, below, Kat announces each step of the game-turn completion (L1-4).
Extract 7: 160712 Tash Kalar 2_0:9:34 The announcement of her final game action (L4) "draw an extra one [card] for this turn" projects the completion point, and "that's my turn" concludes the game turn, making the completion point available verbally. The next player, Adam, however, initiates repair, checking whether Kat had completed the necessary two game actions for her game turn and treating it as unclear. Such repairs must be done no later than the transition space, as once Adam begins his game turn, Kat can see what Adam's own game actions would be and so she is no longer fairly able to change her prior game turn. This extract demonstrates that reaching a completion point and transitioning to the next player are accomplishments, requiring acceptance by all players, which was done tacitly in the previous extracts.
In everyday interaction, given the orientation to each speaker having rights to one full TCU (Sacks et al., 1974), for a speaker to lapse (Hoey, 2015) halfway through a clause would be cause for repairthe speaker would not have completed the TCU and may have left important elements of the TCU for subsequent action unaccomplished. With games, it is always the case that an unfinished game turn causes problems for subsequent game turns, as it undermines the central means by which gaming itself is accomplished as an activity. Thus, if the practices for reaching a completion point are deemed insufficient, players initiate repair. Extract 7 shows repairing the conditions for pre-possible; Adam is not repairing whether Kat believes her game turn to be finished, as this is evident in her utterance, but is repairing whether her actions have sufficiently accomplished a game turn, for the purposes of what the players are willing to accept in this game, in any case (Liberman, 2013). In contrast, below, Tina repairs whether Hal has reached the completion point (L20), rather than repairing whether Hal had completed sufficient game actions. It is evident from the preceding moments that Hal has accomplished game actions that serve as pre-possible completion-specifically, purchasing a "building" and making a new one in the deck available for others to purchase. As the building's features are hard to read, players discuss it aloud (L3-9), which temporarily suspends the game-turn progression. When the joint attention to the new building ceases (L17), Tina initiates a confirmation check (L20) that the game may now progress. Tina is the next player.
Extract 8: GN Lords of Waterdeep_56.16 Possibly due to the discussion of the newly available "building," Hal does not withdraw from the playing space. He does occupy himself with arranging his materials (L15) and looking through his cards (L19), practices that are typically done after having withdrawn from the playing space and after the next player has begun his or her game turn. Tina monitors Hal (L14-22), so it is clear she is attentive to his motions, rather than distracted by the new tile. Regardless of whether Tina is actually uncertain whether Hal has any remaining game actions for his game turn to do, by initiating a confirmation check Tina reorients joint attention of all players to game-turn progression. She provides a warrant for herself to begin to play, in the absence of a clear indication from Hal that he is at a possible completion point. She chooses to avoid potentially overlapping with the end of Hal's game turn.
If, despite game-turn ending practices, the next player does not start, additional methods can be used, including prompting with an order. Here, we see Rob finishing the pre-possible completion elements of his game turn (L1-4), and then restricting his access to the playing space by both clasping his hands on the table edge (L4, making them unavailable to move pieces) and then withdrawing from the table (L6). Despite all this, Stefan does not begin the next game turn. Rob thus prompts him to play (L9): Extract 9: DT Small World Sky Islands_20:24 In saying "Go" (L9), Rob makes Stefan accountable for not yet having begun his game turn. Rob has displayed the full series of possible practices for reaching possible completion, even including an utterance, "Yeah okay" (L5), that seems to confirm his decision to himself in a form of self-talk (Keevallik, 2018), as well as further making his game-turn completion visible. In other words, we have reached the post-possible completion environment. In the next section, I will further analyze how players manage progressivity of game turns once possible completion has been passed.

Post-possible completion: Accounting for delay
In everyday interaction, the post-possible completion space is for expanding a completed turn and/or orienting to a completed turn (Schegloff, 1996, p. 90). A game turn cannot be expanded, at least, without extensive accounting and discussion of whether "take backs" are acceptable (see Liberman, 2013, p. 108). However, players use the post-possible completion space for a different purpose: to prompt the next player to begin their game turn. Several points of evidence demonstrate that these are prompts and not next player selections: the preallocated game-turn order is available to all players (often in visible form on the table), players do not name the next player unless the next player is displaying evidence of disattention or delay, and players treat disattention and delay as accountable.
For example, in Extract 10, Sarah fails to begin her game turn in a timely fashion (L4-5), and Mary prompts Sarah as the next player (L6-8).
Extract 10: RPS Coup 170404_37.00 That Sarah is the next player is available to all present; she is tacitly selected (Lerner, 2003), and the group has gone through the counter-clockwise round of players several times already. Mary withholds prompting Sarah until Will initiates these withdrawal motions (with a point, L7), showing again the orientation to withdrawal as the point of completion. Mary is not merely naming or selecting Sarah, but prompting Sarah to play, and calling her to account for not yet beginning her game turn. It is a pursuit (Button & Casey, 1985). Sarah is given opportunity to prebegin her game turn (before the prompting) throughout Will's lengthy rearranging of his pieces and withdrawal (L5-6). However, Sarah displays no evidence that she is aware of or is starting her game turn (or that Will's has finished, L6); in fact her gaze is down, her cards are untouched and unreadable on the table, and her hands are off the table, displaying no incipient playership. Mary's prompt also shows sensitivity to the calling-to-account nature of her action, as she begins with pointing (which Sarah might not even see as she begins to gaze around the table and potentially notice it is now her turn, L7-8), before verbally reminding Sarah that it is her game turn (L9). One is morally accountable for knowledge of the player order (Stivers et al., 2011), for keeping track of whose current game turn it is, and for giving timely response to game-turn transition.
Prompting occurs only after possible completion. It generally takes the format of describing who is next, such as by their name (Extract 11), color of pieces (Extract 12), character, or with "you're up," "your go," "your turn," though prompting can also take the form of a directive ("go"). Prompting is one option for the postcompletion space, in that it orients to and pursues the fact that the next player or game turn has not yet begun (increments especially do this in English, Ford et al., 2002). Players are then accountable for timely entry. Even if, according to their preallocated rights to take a game turn, players could in principle take however long they like, in any actual game players are beholden to Garfinkel's (1963, p. 200) observation that no player or game "permits the time of occurrence, duration, and phasing of moves to be defined entirely as a matter of the player's preference". 3 Literally taking however long they like to play would likely result in abandoning the game as it stands, and taking a long time raises questions about attention, commitment to the activity, capacity to play, and so on. What happens, then, if a player is not yet ready to play? The prior game turn may have changed the options available for the current game turn, altering plans or strategies. How does a player delay appropriately? The technique to accomplish this is a display of "thinking" (Hofstetter, in press). Players engage in displays of attention and awareness to the game turn and of incipient-but-notyet-ready action. These practices display for other players that the current player is aware of the game- turn progression (so needs no prompting) and is actively orienting to progressivity but faces some barrier to completion. Here, Hal completes his game turn (L1-2) and withdraws (L3-5). After 0.6 seconds of silence, he adds a postcompletion "Done" (L6), reclosing and emphasizing the completion of his game turn, possibly since the next player, Tina, is not yet collecting her pieces to begin.
Extract 13: GN Lords of Waterdeep_1.43.08 Tina demonstrates her active attention to the game through a turn-starting clicks (Ogden, 2013) (L8), leaning to get a better view of parts of the board (L9-10), and self-talk concerning possible options (L12). These practices also project incipient action, by displaying active efforts to move the game turn forward. Hal returns to organizing his own pieces (L14), which enacts willingness to wait for Tina to play (rather than asking for her attention or action; it does not hold Tina accountable for not yet playing). Tina can thus delay actually playing but still begin her game turn.
Thinking practices are so connected to prebeginning a game turn that, if used at the inappropriate time, may cause a disruption to intersubjectivity concerning whose game turn it currently is. For instance, according to the game-turn order below, it is actually Bird's game turn, but they are not displaying thinking (L1). Spider is the next player, and they begin to display thinking practices instead-"uhm"-ing (L2), singing (L2), and leaning in toward the board (L4-5). Stevanovic (2013) has also pointed out this connection between humming and displayed unavailability; "thinking" practices like singing display unavailability to the other players due to attention to the game turn.
Extract 14: 160908 Dominant Species 1_19:28 Bird treats Spider's thinking practices as a display of starting a game turn and initiates repair as to whose game turn it actually is (L6). Only once the players all confirm it is actually Bird's game turn does Bird start to .ffff engage in thinking practices (L14-16). In this way, Bird rectifies what was absent previously-a display of incipient playership. Spider likewise ceases doing a display of "thinking" now that it has been discovered to be inapposite. In this section, I have analyzed how players manage the post-possible completion space. If the next player fails to begin the next game turn in a timely manner, then other players may make them accountable by prompting them to play. Prompts are not selections, given the preallocated game-turn order but are sanctions for not displaying timely attention to the game's progress. If next players want to avoid such accountability but are yet unable to play immediately, they may engage in thinking practices whereby they demonstrate active attention to the game turn, ponder aloud their choices, and demonstrate active efforts to play incipiently.

Overlap and skipping
I have demonstrated above how players orient to similar structures in game-turn transitions as in Schegloff's (1996) analysis of everyday turn transitions, with orientations to possible completions points and the pre-and post-space around them. I have shown instances where prebeginnings were treated as acceptable, as long as the overlap occurred after the pre-possible completion point (see Extract 1, 3). The final question to address is how players deal with and what they treat as (interjacent) overlap (Jefferson, 1986)-that is, a player taking a game turn when it is not their moment to play.
Players are accountable for keeping to the game-turn order, and skipping game turns is a sanctionable occurrence (though players manage without a mediator, contrast to teacher mediation in Hazel & Mortensen, 2017). When Cat's game turn reaches possible completion, below (L1-4), the next player should be Bee, but Ada initiates her game turn instead (L7) in overlap with Cat's post-completion assessment of her own game turn (L5-6).
Extract 15: 170322 RPS Coup_26:42 [heh Ada is resoundingly sanctioned with yelled corrections from both Bee and Cat (L15, 16) and retroactively transforms her game turn into a preannouncement of what she will eventually do (L19), providing a joking sort of account for her actions. This demonstrates how quickly and how dramatically players will hold each other accountable for maintaining proper game-turn order.
With respect to overlap, I have described above (see Extracts 2-3) how players avoid playing any earlier than pre-possible completion, which is reached when the current player completes the last (set of) game action(s) on his or her game turn. One reason for this is that players are waiting for the moment when the current game actions are "set in stone"-that is, when the current player may no longer unaccountably change their game actions. The further in time (and in game-action sequences) a current player gets from making game-action choices, the more accountable they are for making alterations. As already seen in Extract 1, the space between pre-possible completion and the possible completion point is one of the last places for self-repair, without being made accountable by other players. In contrast, below, players orient to the accountability of changing the game turn after their game turn and the next game turns had passed. Players orient to the necessity of asking permission and to consensus, before making any change. In Extract 16, prior player Bird builds up both a preaccount for needing to change their prior game turn (L5-11), as well as adding subsequent accounting in the absence of response (L17). Bird treats it as necessary to get permission-in particular permission from Spider-and does not act until that has been given. In Extract 17, more time has passed since the game turn the player desires to change. Kat treats herself as even more accountable than Bird, by framing her request even more strongly to prefer a refusal (L1) and mitigating the accountability of a refusal with "It's fine if the answer's #no::.#" (L8).
Reptile puts into members' terms one reason for avoiding post-game-turn change: it is unfair because players make choices based on the information available at the time. However, this also provides insight into why players avoid playing until after pre-possible completion is evident; overly early play would provide the current player with information about the next player's game turn that they (the current player) should not have while making their own decisions. In other words, a current player should make strategic decisions without being able to know the future; knowing the next player's choices gives the current player an unfair advantage, which undermines fair play, a valued element of gaming as an activity. This means that a next player should not play until the prior player is unable to change their (current) game turn or, rather, when they become accountable for doing so-when the current player is past the point of self-repair.
Returning to the transition space, we will see that the precise moment when such "inability" to change decisions is a matter for members to negotiate in situ. In the game below, game turns involve placing a token, taking the resources (e.g., "money," "wizards") that placement enables and then, optionally, playing a "quest." The quest involves paying a specified set of collected resources back to the bank and collecting points in return. Typically, next players wait until such a quest is completed (players usually do one every game turn) or the opportunity explicitly passed up. However, the information next players need for choosing their game action-which location to place a token upon-is available as soon as the current player begins dealing with the quest, because the decision concerning which location to visit has been made. The decision on location is almost never repaired, because the resources collected at that location are typically used immediately for the quest at hand. This means that it is possible to interpret the start of prepossible completion as beginning as soon as a player has withdrawn from the token placement, rather than once a quest has been completed. Such an interpretation is made below, by Brandon starting his next game turn in overlap with Tina's current game turn (L8). Tina makes Brandon aware that she is (a) not foregoing the optional quest and (b) still dealing with it, by ordering him to "wait" (L11-13). Brandon at first withholds playing and apologizes (L16-19). However, he switches to accounting for his actions as logical and acceptable, treating Tina's game actions as now unchangeable (L19-25). Although at least one other player treats this as acceptable (L23-26), Brandon himself then pauses his game play again (L29-30). At this point, he has taken his first game action (L21), which he can do as long as he treats Tina's game turn as unchangeable, but he cannot continue with taking his own quest until Tina's is finished, due to Tina's quest's possible consequences. This sequence demonstrates an extreme instance of overlap and how it is made sense of as not interjacent. It also demonstrates how the pre-possible completion space is a negotiable, member-organized phenomenon, in the sense of where it starts but also in the sense of what is accountable within that space. The preallocation of game turns results in orientations to whether a game turn is sufficiently unchangeable to permit next players' game turns to progress, the alternative being treated inevitably as early and interjacent, as well as unfair.

Discussion
In this paper, I have analyzed the methods board gamers use for managing game-turn transitions in a round-based, preallocated turn-taking system. This analysis demonstrated how players actively organized game play and how game-turn transition was coordinated as part of achieving gameplay itself. I have demonstrated that players organize game-turn transitions analogously to Schegloff's (1996) system for turn transition, in that players orient to pre-possible completion, possible completion points, and postpossible completion additions. First, pre-possible completion was shown to depend on completing preallocated elements of the game (specific game actions) and that players used additional practices to then reach possible completion, such as withdrawing from the playing space. Second, players used the postcompletion space to prompt (when necessary), rather than select, next players, treating selection as what ought to be known already by all players. Third, although players oriented to accountably beginning their game turn on time, sometimes they had to begin but not actually play. Thinking practices, which demonstrated attention to the game turn and active efforts to progress it, solved this problem (see also Hofstetter, in press). Most critically for the analysis of settings involving preallocation, transitions were shown to be a member-organized phenomenon, not a passive adherence to preallocated features. The preallocated features, such as the sequence of game actions, make an inferential framework (Levinson, 1992) on which to project completion, much as prosody, syntax, and embodiment provide resources for projecting TCU endings in everyday talk. A relevant next step would be to analyze the means by which such a framework is established and begun, in situ, which would be relevant for all preallocated systems.
The preallocated nature of the turns provides additional affordances that everyday talk does not. For instance, as players cannot (legitimately) progress the game in the absence of a game turn from the current player, practices for managing ongoing delay arise. In doing so, players uphold the "fairness" (Hofstetter & Robles, 2019) of the game-specifically, in maintaining equal access to game turns for all players. The preallocation both permits and enacts an ideology of fair play, something that may be worth investigating where institutional turn systems are similarly concerned, such as judicial settings.
A further next step is to analyze non-round-based games, and games in other cultures and languages, since they are such a ubiquitous human phenomenon (De Voogt, 2017;Finkel, 2007). Preliminary evidence in Zinken (2016) and Sutinen (2014) suggest the proposed system here applies at least in Polish and Finnish. Furthermore, an analysis of the practices for creating and maintaining game-turn order and the sequence of events involved in achieving a game turn is necessary.

Notes
1. I am very grateful to an anonymous reviewer for suggesting I make this a primary comparison point. 2. Although across the corpus, players largely use the "touch rule" common in chess, it is only in professional chess that such a rule was ever seen to be enforced. The touch rule stipulates that once a player releases hold on a token, they have committed to that token's placement for that game turn. However, even in competitive chess players often ignore the touch rule to fix pieces that mistakenly fell over due to haste of placement. 3. Thank you to the same anonymous reviewer for suggesting this connection.