It takes a special education to teach persons to know what they don't know

(Sacks 1989, 127)




This is a collection of terms that derive from a series of fourteen lectures given by Harvey Sacks in 1964 and 1965. Sacks is famous as the originator of conversation analysis (CA) which is a method of analyzing recording or naturally occurring spoken conversations The terms collected in this document have been gathered as a part of my own research into Sacks and more broadly the methodology of (CA). I don’t pretend the list is definitive or even particularly exhaustive. Others may also take issue with the ways I have assembled the various fragments found here which respects neither the chronological development of Sacks’s thought not the particular context of the lectures in which these insights were offered. One of the motivations for offering this document is to address the objection of many critics of CA, that it is a method which is obsessed with triviality. I think these lectures show how those criticisms are misinformed and that Sacks (although dealing with what he himself called ‘microsociology’) is nevertheless concerned with tackling some profound and wide-ranging themes that concern sociology per se. I would also like to draw attention to the elegance and originality of Sacks’s thought, particular I would point to his remarks about ‘glances as actions’, which is a good place to start for anyone who wishes to get a handle on Sacks’s though (although ironically this section has the least to do with what one might ostensibly define as ‘talk’). In addition his remarks on ‘measurings’, ‘jokes’ and ‘proverbs’ name devices which Sacks claims function to regulate and control the ordering of conversations by interlocutors and in that respect can be seen a form of social construction in which members and the societies of which they are a part are constituted. A final justification for making this available online is that Sacks’s books are prohibitively expensive, and I hope by publishing these notes that I may play a small part in bringing his thought to the attention of a wider audience.

As to the issue of copyright, I do not own the copyright on these lectures. They have been taken from the journal Human Sciences.

Citation: Sacks H (1989) ‘Harvey Sacks Lectures 1964-1965’. Jefferson G. (Ed.). Human Studies Vol 12 No. 3/4. Netherlands: Kluwer Academic Publishers.

Blackwell publishing also have some claim of ownership on this material (as they have published a complete volume of the lectures, edited by Gail Jefferson). Readers are advised that the quotations that appear here, do so under the normal conditions of scholarship and educational free use and should not be copied or distributed.

Rod Munday


Lecture One:
Rules of Conversational Sequence (pp. 35-45)

Lecture Two:
On Suicide Threats Getting Laughed Off (pp. 53-61)

Lecture Three:
The Correction-Invitation Device (pp. 65-70)

Lecture Four:
An Impromptu Survey of the Literature (pp. 71-77)

Lecture Five:
Suicide as a Device for Discovering If Anybody Cares (pp. 79-88)

Lecture Six:
The M.I.R. Membership Categorization Device (pp. 89-99)

Lecture Seven:
On Questions (pp. 101-110)

Lecture Eight:
On Measuring (pp. 113-123)

Lecture Nine:
"I Am Nothing" (pp. 131-137)

Lecture Ten:
Accountable Actions (pp. 139-148)

Lecture Eleven:
On Exchanging Glances (pp. 151-166)

Lecture Twelve:
Sequencing: Utterances, Jokes, and Questions (pp. 169-179)

Lecture Thirteen:
On Proverbs (pp. 183-193)

Lecture Fourteen:
The Inference-Making Machine (pp. 197-211)



Key to text

For the most part I have tried to let Sacks speak for himself. Direct quotations are written in 'Times New Roman' . I have provided page references for the quotations, given before the quoted text. Readers are welcome to work out which lecture they came from by referring to the contents list reproduced above. Note that some gaps in page numbers are because fragments of other talks have been inserted between them in the form of exerts. In the text itself, the use of square brackets "[ ]" denote sections text that has been inserted by me to smooth over the discontinuities created by the editing process. Remarks inside rounded brackets that are written in the sans serif 'Arial' font identify my own commentary. Such commentary is reserved mainly for explicating terms such as ‘class’ and ‘members’ the meaning of which is not explicit stated by Sacks but becomes apparent as you read through the lectures. I hope that my paraphrasing captures something of Sacks’s intended meaning. Finally, note that the reason for Sacks’s preoccupation with suicide helplines is because he studied collection of recordings of conversations collected from the Los Angeles Suicide Prevention Center. For Sacks, these recordings were adequate research objects because they represented examples of naturally occurring talk that were not ‘contaminated’ by other social science methodological procedures.

* * *


Harvey Sacks - Lectures 1964-1965


A-B reduplicated [169] the basic sequencing format for conversations … One party talks, then the other party talks, then the first party talks again, etcetera … the term 'two party' … does not necessarily mean two persons. The 'two-party' conversation may be a basic format such that conversations having more than two persons present can take a two-party form. That would involve persons dividing themselves up into teams of a sort, and alternating according to team membership… [170] if A-B Reduplicated is the format of [171] conversations, then there is no specific length that a conversation takes, [and] there can be enormous variance between two conversations as to how much was said… It's not a situation where persons have to monitor how much they've talked as compared to how much the other has talked, to find that the conversation can, or ought to, close. Thus… we could say that we have a 'minimal conversation', "Hello," "Hello." And we could say that if at least that took place, then a conversation occurred. (See also conversation; conversation analysis; sequencing)

[absent ‘I’] [79] Those usages, where a person says "you" or "one" (or people) as a way of stating something that they propose thereby to be a generally correct remark, and how they are defended, and what kind of attacks they can be subjected to, is something we can watch. [141] this Jewish fellow says, "It's not that they did it to me, they don't know me, and I don't feel personally affronted. But that they treat people that way!

accountable action [37] "Why?" is a way of asking for an account. if the account is to control the action, then if you can find a way that the account controls the alternative action than it proposed to control… [35] exchanges occur as 'units'. That is, "Hello" "Hello" is a unit, and "This is Mr. Smith may I help you" "Yes, this is Mr. Brown" is a unit. They come in pairs … [174] With "Hello", "Hello" and things like it - members of a class of paired activities such that if A uses one, B's proper move is to use one also. [36] Saying "This is Mr. Smith may I help you" thereby provides a 'slot' to the other wherein they properly would answer "Yes, this is Mr. Brown." [37] It seems to be quite important, then, who it is that offers the account. If they speak first, they have the right to choose their form. [and then it is] the task of the person who is offered the account… to, in some way, counter it. [36] [if] there is a rule that the person who goes first can choose their form of address and thereby choose the other's, then for the unit, "This is Mr. Smith may I help you" "Yes, this is Mr. Brown", if a person uses "This is Mr. Smith..." they have a way of asking for the other's name - without, however, asking the question, "What is your name?" [37] The task of the person who is offered the account can be then to, in some way, counter it. Where, alternatively, persons who offer an account seem to feel that they're somehow committed to it, and if it turns out to be, for example, inadequate, then they have to stand by it The fact that you could use questions - like "Why?" - to generate accounts, and then use accounts to control activities, can be marked down as, I think, one of the greatest discoveries in Western civilization. It may well be that that is what Socrates discovered. With his Dialectic he found a set of procedures by which this thing, which was not used systematically, could become a systematic device. [140] Phone calls seem to be 'accountable actions'. By that I mean that by and large on the first opportunity to talk after greetings, the person who's called gives an account of how they happened to make the call. We can also notice that they are 'symmetrically accountable'. That is, if somebody calls [a helpline then they] give an account [and if the helpline calls somebody then] they give an account … Persons who call up someone they are not intimate with, often construct accounts of how that came about... [141] Now, (in the context of ) getting help, there is a sense in which the two sorts of accounts are substitutable. You can use one or the other, and either of them can work. But I want to notice that the use of each of these has a different source... To call an organization and propose that the call takes place by virtue of the operation of some other organization, involves informing those you're now calling that this action is not the first step in the search. That's non-trivial because persons tend to start a search for help with that place about which they have the most rights to have an expectation that they'll get help. That's very unfortunate in a way, because if they don't get it there, they may apply a formula to that fact: "If not there, where I had most rights to expect help, then where can I expect help from?" That is, they treat the sequence of calls as informing them about the likelihood that any next call will be a success. The other sort of account; those in which persons propose straight out that they need help. first thing that seems to be worth considering is how asking for help is regulated. In an earlier discussion I drew a distinction between two classes of 'others', (see insiders and outsiders) and said something to the effect that one class were persons with respect to whom there was such a bond of obligation that you could directly say to them "I need help"-, for example, 'family' … When somebody calls up an agency and says "I need help" - which is the proper thing to say when you're talking to an insider - a signal ought to go up that something is going on here. What can be present is, they've turned to an insider and gotten no help, or for some reason they find that they're not able to turn to an insider. That is [why when] we get callers starting off with 'requests for information' which involve, then, that over the course of the conversation help will be offered without being 'asked for'. Because asking for help is something that one doesn't do with respect to strangers. So, "I need help" and "How does this organization work?" seem to be alternatives in the sense that persons who use them have perhaps come to do this call by the same paths. (See also Correction-Invitation Device; dependence; entitlement; obligated others)

accountable actions avoidance [35]

A: This is Mr Smith may I help you

B: I can't hear you.

A: This is Mr Smith,

B: Smith.

[38] When you say "I can't hear you", you provide that the other person can repeat what they said. Now what does that repetition do for you? Imagine you're in a game. One of the questions relevant to the game would be, is there a way in that game of skipping a move? It seems that something like "I can't hear you" can do such a job. It is not simply that the caller ignores what they properly ought to do, but something rather more exquisite. That is, they have ways of providing that the place where the return name fits is never opened. The rules of etiquette - if you want. (See also accountable actions; A3N; jokes)

Adam’s problem see subversion (Adam’s problem)

adequate complete utterance [170] something that a person can say, which, upon its completion, provides for the relevance of the *sequencing rules. A sentence may be complete, and one could tell that it's complete, but that wouldn't tell you that the person is finished speaking for now. But if they use an 'adequate complete utterance', then, by virtue of the fact that that unit is complete, the sequencing rules are relevant… [173] the ways one shows that one has done something which is an adequate complete utterance (appropriate for the use of the sequencing rules) is to repeat it. So I have these reports where a child says "Hi," there's no answer, and the child says again, "Hi!" And then there's a "Hi" in return, and the child will take that as having been sufficient. (in modern CA, this an ‘adequate complete utterance’ is known as a TCU or Turn Completion Unit)

alienation [165] (An acute lack of confidence is your membership of a class so that you cannot judge what is normal) one of the senses of alienation would presumably be that you would feel tremendous doubts about doing (things like exchanging glances with others as a tacit means of communicating what is normal between you). And one would have tremendous doubts about what those smiles and disapproving glances are about; what they're looking at you for. Schizophrenics are always reporting that people are looking into them, talking about them, know what they're thinking. And you can see that they haven't constructed a machine which produces events going on in the world, wholly out of their heads.

anti MIR devices see tautological statements (as anti MIR devices)

A3N device (Account Apparently Appropriate Negativer) [68] Where persons are engaged in trying to get an account from somebody, there's an object that the person who's being questioned can slip in [which] cuts off the basis for the search for an account. For example, the line, "Everyone does, don't they?" is an A3N… when such general-purpose devices as A3Ns and *proverbs are used, others don't attempt to question them or contradict them. I think there's some reason why we don't much see attempts to question or contradict these things, and that is that they may be such basic objects - that is to say, Members are so committed to their correctness - that if you undercut one, exactly what you've undercut is not clear. And one doesn't know exactly how we can continue talking. [69] A woman was collecting research materials by going into parks with her children and just starting conversations with people. One of the things she reported was how the conversations began … there would be a woman sitting on a bench. This woman would go over to the bench with one of her children, and sit down. The little boy would wander around for awhile, then he'd come up to her and she'd say, "Go away, I [70] want to sit and rest." Sometimes he'd go away, but sometimes he'd sit there, annoyingly. And then the other woman would turn to her and say, "They're all like that, aren't they." And she'd say "Yeah" and they'd get into a conversation. I asked her, "Did you ever say no, or something like that?" And she said "Yeah, when I first got out of college I was all full of information. People would say that to me and I'd say 'Well I don't know, my kids aren't.' And they'd always stop talking right then and there." (See also accountable actions)

[A3N variant] [79] [In telephone conversations] you might look at the way [a] caller sets up giving her answer -by the use of "Well it's the same old childish reason that everybody wants to commit suicide" - and compare it to the *A3N device. That is, the A3N can provide that an account need not be produced. The sort of line this woman uses might provide that the account she is about to produce is not challengeable, needs no defending. (See also A3N device)

average [159] [Sacks’s students found that] persons [160] didn't exchange glances over an average person. Now what you have to attend is the notion of 'average' that you employ. Because I don't think it's the case that you use[in the sense of constructing] a distribution over that set of persons and provide that those that stand out from that set, after let's say, watching awhile, are the unaverage ones. Rather, somebody's being outstanding may be quite irrelevant to the collection of persons surrounding them in the scene; that is to say, those who have passed before or who come after. [Similarly, when people are considering] odd suicides,… what you have is that an odd suicide gets separated off and is not used to consider the, quote, normal ones. So a stream of odd ones is only something to be remarked upon, like "Isn't it amazing today, there's such a stream of absolutely beautiful women who walked through such-and-such a place." It isn't that one then finds that one is going to modify the notion of 'average' because of that. And that then means that those categories have to be given special attention. (See also decomposable; normal)

base environment [40] We can consider [*composites] in terms of what I'll call the 'base environment' of [their] use. By 'base environment' I mean… one way of informing you what kind of a place it is. So, if a new institution is being set up, then there are available in the society whole sets of ways that persons go about beginning conversations, and one could, for example, adopt one or another of a series of them as the ones that are going to be used in this place [41] ….As a general matter, then, one can begin to look for kinds of objects that have a base environment, that, when they get used in that environment perform a rather simple task, but that can be used in quite different environments to do quite other tasks. So, a matter like 'credentials' can be handled by [a composite like a] "May I help you" device. There will be lots of other devices which have a base environment, which do some other tasks in some other environment. If the base environment is something like a department store, then [the meaning of "May I help you" will be different to] when it's used in other places - for example, a psychiatric hospital (See also constructives)

categories [91] All the sociology we read is unanalytic, in the sense that [sociologists] simply put some category in. They may make sense to us in doing that, but they're doing it simply as another *Member [of their society]. They haven't described the phenomena they're seeking to describe - or that they ought to be seeking to describe. What they need to do is give us some procedure for choosing that category which is used to present some piece of information. And that brings us back to the question, are there procedures that Members have for selecting categories? One of my aims is to show that there are… The way you get a piece of knowledge involves pulling out the name and putting in some category. Then one gets, not 'John did X', but 'a such-and-such did X'. In that way one gets additions to any given body of knowledge about such categories. And what we find is that an enormous amount of what we could call the lay theories of social actions are fitted onto these categories. (See also M.I.R. category device; M.I.R. category device modifier)

category set [89] A set which is made up of a group of categories.

ceremonials [55] (Normative bits of talk consisting of stock exchanges) for example, "How are you feeling?" [is] routinely used between persons as either a greeting or greeting substitute. And it's used between persons who needn't have very much intimacy. "How are you feeling?" to which you return "Fine." If one person, then, uses a ceremonial, the other properly returns with a ceremonial. [56] I've sat around in hospitals, and in a hospital, persons who are, say, recuperating from serious diseases may be sitting in wheelchairs outside their room or in the commons room, etcetera. A doctor walks by a person who looks like they're just about to go, and says "How are you feeling?" and they say "Fine." Sometimes, however, a person may take that "How are you feeling?" and attempt to use it to present their troubles. And one sort of thing that happens in that case is that persons who listen when somebody begins to tell them their troubles, talk about themselves routinely as 'soft-hearted', 'fools', and that sort of thing. And when persons talk about themselves as soft-hearted with respect to [*obligated] others… they listen, then they find themselves 'involved'. Involved, however, without the basic properties that would initiate their relevant obligation, but not knowing what to do. And not knowing how to get out, either, because they 'know too much.' [60] One of the things that's reported about persons who have to deal with [people suffering from mental illness such as] paranoids is that they feel weak, experience a terrific lack of control when they encounter them. [For instance they ask "Hope you have a good time" and get in reply "why?"] Now you could go about trying to examine that, perhaps by studying let's say the comparative dynamics of the persons, or various other things. But I think you can get an idea of how they would have that feeling of weakness by just examining an interchange like this. We're talking about ceremonials. The normal answer to this "Hope you have a good time" is "Thank you". And if one uses a line like "Hope you have a good time" one can expect to control the return of the other. In this case the line doesn't control the return of the other, and we can at least begin to see what it means to feel weak: Having an expectation of doing something as controlling, and finding out that it isn't that at all… But furthermore, this "Why?"-return casts this "Hope you have a good time" into the character of an 'accountable *act[ion]. (Asking the question "why", for instance, turns a ceremonial into an *accountable action). Doing something that wasn't seen as accountable, having it turned into something that is accountable, one doesn't have an account. One offers, then, an account which one feels is quite feeble. It's feeble. [56] persons who are causally bound are obliged to give help when help is asked for. That means in part that they're in bad shape if they don't give help and trouble occurs. They're responsible for someone. Others hold them responsible, and they feel responsible. The question is, is there some way that they can go about refusing to give help without 'refusing', in the same way that I've talked about refusing to give one's name without [57] 'refusing'? One solution would be to find a way to set up the first remark as the first remark of a ceremonial. Because then the proper return is a ceremonial. Three common classes are *jokes, games, performances. They all have the character that the next move - or some other given move in the sequence - is the end of it, and that's the end of the whole thing. When somebody says "I'm going to kill myself, if the other can cast it into one of the ceremonial forms, then that can end the interchange. One wouldn't have heard the 'cry for help'. One would have heard a joke. And one would have behaved properly with respect to a joke. And it appears that, alternatively to giving help, one gets cases of just those three common classes of ceremonials. Somebody laughs, or they say "Nice performance", or "Quit playing". And that would provide, then, for closing that thing off without, however, having been in the situation of refusing help in the sense of saying 'no', or other such things. So we can see how that form provides for this thing to happen. We can note that there are classes of events which, between persons who are not terribly intimate, get initiated via ceremonials. "Would you like to come over for dinner tonight?" "Sure." That is, for these kinds of events to occur, there has to be an invitation, an offer of some sort. So that's one task of ceremonials - they do the job of providing for these events to take place. They do another job, in a way. When persons are quite intimate, then one way they measure that is by virtue of the fact that invitations are no longer relevant. You can go over to their house without being invited. And people will say to each other, "Come over any time you want." Now with a husband and wife, one gets a version of this not feeling wanted, which goes something like this:

Wife: Why don't you ever ask me to go out to dinner anymore?

Husband: If you want to go out to dinner why don't you just say so?

Wife: I don't want to go out, I just want you to ask me.

What she's picking up here is the absence of ceremonials. And ceremonials have this double use. On the one hand they are properly used to provide for persons to do things - come over, go out to dinner, etcetera - at some state of a relationship. At another state those things happen without them. And they're not absent. Indeed, it surely happens that somebody might say, "Well why don't you come over tonight?" and the [60] other says "Why are you suddenly making a big deal of it?" But this double use then provides that when somebody has some doubts of some sort, they could focus right in there; that they see this thing is absent, and see the absence via the position of one who is not in the position of intimacy. And they don't know quite how to handle that matter. Because if they complain, they get "Why are you standing on ceremony?" and if they don't complain and don't get the invitations, they figure "Jeez, it's the case that I'm not wanted anymore." [56] On the other hand, the fact that there is that ceremonial relation between "How are you feeling?" and "Fine", may set up the following situation… between [a] psychiatrists and patient:

A: How are you feeling?

B: It's a long story.

A: That's alright, I have time.

What is this "It's a long story", and things like it, doing here? The person knows that the line "How are you feeling?" is a ceremonial line, and it's a breach of the proper forms to begin to launch right then and there into what it is that's bothering you. So what they then do is try to initiate another ceremony which would provide the basis for them talking. Typically this other ceremony is nicely done, in that what one does is offer a tentative refusal, like "It's a long story" or "It'll take hours", which turns it back to the other, referring to their circumstances; for example, their schedule. And it invites the other to then say that their schedule does not control your activities, so go ahead and talk. (See also obligated others)

class [157] "[T]he class permits you to see what it is that's there."… [They are] the explanation that opens our eyes. So, the classes and their categories permit you to see. (The notion of ‘class’ is a central idea in that underpins most of Sacks’s approach to conversation analysis and his philosophy. Because of its centrality and because his use of it is so multifaceted I think that the notion of ‘class’ in the Sacksian sense is best approached obliquely firstly by considering the following quotation):

[151] An assignment has been given, having to do with observing people exchanging glances.

Q: Can I ask a question about this assignment.

A: Yes.

Q: For each person that we notice looking over at somebody are we supposed to ask them if they know the person?

A: No! Don't ask!

Q: But we have to write down the class and everything, how can we possibly know if we don't ask.

A: Yeah, class membership doesn't mean Junior,


Q: Oh I realize that.

A: I don't mean social class, either. I mean class in the

sense that I've been talking about class. Any class. Whatever it might be. You figure it out. When two persons exchange glances, see if you see anything similar between them, and see if you see what it is that might be what they're noticing. If you find that you don't know at all, you can say that. I don't think you're that naive. You walk through the streets and you're constantly classifying the persons you see.

(Class is a term that is used in two primary senses by Sacks. The first is to name the things that one can discover by analyzing conversations, that is to say the perception of patterns or regularities in conversational exchanges). (See also accountable action; A3N; categories; forms of address) (These in turn reveal classifications where class used in its second sense as designating the perception of patterns or regularities held by others which can be seen as explaining their behaviour in social situations. Norms are classes of patterned behaviour which are performed either consciously or tacitly by members of a give social group given that certain base environmental conditions are met and they have the function of warranting a persons sense of entitlement or belonging to a group or society). (See also alienation; average; decomposable; entitlement; glances are actions; measurings; Member; M.I.R. device category. Compare Class I and Class II rules)

Class I and Class II rules [145] it's rather well known that very young children have, from the perspective of adults, a rather poor notion of causation. They don't know how things happen to happen. Now, among the ways that adults go about formulating rules for children, are two which it's important to distinguish. Call them Class I and Class II [rules]. A prototype of Class I is, "Don't stick your hand on the stove." Prototypic of Class II is "Honour thy father and mother"... With respect to an adult's conception of reality we would say that these two are different, in that for Class I the consequences, whatever they are, naturally flow from the act done. If you stick your hand in the fire, you get burned. Whereas for Class II, that's not so.. For a lot of things that you do that are said to be wrong or harmful, somebody has to do something to you for you to get the negative consequence. You can 'get away with' things of the Class II sort. Now it's supposed, and it seems reasonable, that there's a stage when children don't know the difference between those two classes. That fact is very important for adults because they exploit it heavily. The way they exploit it is, they'll formulate a whole range of what, as adults, we would talk about as Class II-type rules, in terms of Class I operations… So that parents say to children while given them a spanking, "I don't want to do this, it just had to be done," retaining thereby the relevance of Class I. The fact that adults assimilate the one [146] to the other, formulates a very serious set of problems for children. And that is that children are repeatedly faced with the question, "What kind of a rule is that, that I've been told? Class I or Class II?" There's no principled way for them to find out. They have to proceed case by case. And proceeding case by case, they can get into a hell of a lot of trouble. Adults know this and have a whole class of proverbs on it, an instance of which is, "If you tell children not to stick beans in their ears, they stick beans in their ears." … seriously disturbed children are those who go about assimilating the whole range of Class I phenomena to Class II. They then go about checking out the causal properties of the world as though they were normative properties in the sense that Class II rules are. (See also neurotic. Compare class)

common knowledge [67] Sociologists often talk about something called 'common knowledge'. And one question is, what is it that common knowledge consists of? One thing it can consist of is just lists of items that persons know in common. But there are some things it would be nice to know about the phenomenon of common knowledge. One of them is what we could call its 'structural properties'… Also, how it is that what persons know 'in common' is organized. Also, is it the case that the organizational features of what they know 'in common' are also known?... How do they know those classes? For example, do they know them only if you name the class, then they know one or another which are members of the class? What this stuff seems to suggest is that on the one hand they probably do know, to some extent… But they also know them in this fashion: You can name one, and they know, by virtue of the use of that one, what class you're referring to, and can give you another. And that's a non-trivial way of seeing that, and how, common knowledge has its organization seen and understood by Members. [68] And in that regard, another question would be, how substitutable are accounts for each other? Is it the case that one is as good as another? Which ones would be as good as another for this or that account problem? [Would a] *correction-invitation device… not only work for accounts, but for all sort of things; that is, where you can name an item, and get in return another item [?]

composites [41] stock phrases such as "may I help you" which are something like an idiom. "I'm going to call these idiom-like things 'composites'." [42] one of the pieces of information [composites] seem to convey is that whatever it is you propose to do, you do routinely… That is, it's heard as a standardized utterance. (See also base environment; constructives. Compare proverbs)

constructives [41] Ordinary sentences which are understood by taking the pieces of *composites and adding them up in some way. As a composite, "May I help you" is a piece of etiquette, a signal for staling your request - what you want to be helped with. Alternatively, as a constructive, "May I help you" is a question. If one hears it as a question, the piece of etiquette and its work hasn't come up, and "I don't know" is a perfectly proper answer. Further, "I don't know" may be locating a problem which "May I help you" is designed, in the first place, to avoid. In its base environment, for example a department store, it's pretty much the case that for a customer, the question of whether some person "can help" is a matter of the department store having made them the person who does that. [42] one of the pieces of information [composites] seem to convey is that whatever it is you propose to do, you do routinely… That is, it's heard as a standardized utterance. (See also base environment)

conversation [169] For the linguists, almost exclusively the largest unit of investigation, the largest unit they seek to describe, is a sentence… If we want to study natural activities in their natural sequences, we have to deal with, for example, the obvious fact that a sentence is not necessarily a 'complete utterance'… We want to construct some unit which will permit us to study actual activities … in the first place make of it 'a unit' - a natural unit and an analytic unit at the same time? The question then becomes, what do we need, to do that? [171] A conversation could [be said to] take place, if it has an *A-B Reduplicated format… an *'adequate complete utterance', and a 'paired' characteristic. What [172] we're [also] trying to do is find some way of saying, non-trivially, that something is 'absent'. If there were something that was invariably present we would have no trouble. We could say if that thing happens, then 'conversation' occurs. But we need to be able to say that we have a conversation if that thing is present, or if it's absent. The way we can go about doing that is to find that it's always relevant. If it's relevant, then if it's not present, we can say non-trivially that it's not there. And greetings have that sort of relevance, in that there is no rule of exclusion for them. So we can say that *greetings are relevant for any conversation … It's not, then, that we just need "Hello," "Hello" or *members of that *class to have taken place, to have 'a conversation' and to warrant our being able to say that there is a natural analytic unit, 'conversation'. But if we can say about some piece of talk … that it occurred in 'the greeting place', and that piece of talk, whatever it was, provided for the relevance of the *sequencing rules, then we could say that we have 'a conversation'. [173] We can say… that the unit 'conversation' is warranted by the fact that we have at least a minimal thing that's recognizable as 'a conversation'. For it, the sequencing rules are relevant. We can talk about [174] places in it and by virtue of this we can also see that 'greetings', are of some central theoretical importance, though "Hello," "Hello" looks like nothing that one would want to attend to very much. Their consideration does an enormous amount of work for us. And once we're dealing with the fact that we've got sequencing here, and it's regulated, we're no longer in a position where linguistic investigations are usable. Because grammars don't differentiate this way. (See also accountable action)

conversation analysis [197] One of the basic things I want to be able to give you is an aesthetic for social life. By that I mean in part that we should have some sense of where it is deep, and be able to see, and to pose, problems. I'll try to do somewhat more than that. I'll also try to develop a variety of notions of what kind of business sociology is, what its problems look like, what the form of the solutions to those problems are, and perhaps to some extent, some of those solutions… The kind of phenomena we are dealing with are always transcriptions of actual occurrences, in their actual sequence. And I take it our business is to try to construct the machinery that would produce those occurrences. That is, we find and name some objects, and find and name some rules for using those objects, where the rules for using those objects will produce those objects. And we also consider conversation per se, looking at the rules for *sequencing in conversation… [174] examining the sequential building blocks of conversation which are specially relevant in terms of their sequential character. (See also inference-making machine; microscopic sociology)

Correction-Invitation Device [65] Where one wants to get, from the person one is talking to, an account of something - why they did something or why they have something - one way you can do it is by saying "Why?" Another way you can do it is by asking with the name of the class of things you want. For example, a woman is talking to an officer from the juvenile division of the police force. Her fourteen year old daughter hasn't been coming home at night. The woman called the police, the police found the daughter, and now they're talking to the woman. And they say, "Have you ever had this [66] kind of trouble with her?" That is, 'this kind of trouble' is the name of the class. She can then say, "No I haven't had this kind of trouble," she can say "Yes" and then give some instances, or she can say "No I've had other kinds of trouble." Now it also seems that one can ask for an account by naming, in question form, one member of the class, of which the account will be another member. For example, "Is it yours?" She doesn't come back and say just "No … She says "It's Dave's." That is, instead of saying "Whose is it?" which he said earlier but didn't get an answer to, he gives one possibility and thereby elicits, as its correction, another; the actual class member... Now, so far I've talked about the construction of these correction-invitation devices, and said that it's based on the fact that, using a range of classes, you can refer to one member to get another member. (See also accountable action; questions)

cumulative values [132] there is a special class of values, which are what I call 'cumulative'. These have two features: [Either] (1) If you can properly have them at some stage of your life, then you ought to have them at any time thereafter. [Or] (2) You can lose them, so that having them at one stage of your life is no guarantee that you will have them thereafter. For example, the cumulativity of the value 'having children' is recognized in the proverb "Children should bury their fathers." If you had children at some point, you ought still to have them; if they are now dead or if an irrevocable break has occurred... There are other sorts of things which are cumulative, but on which you get a yes-no alternative; for example, being married. For such things, 'no' is apparently equivalent to 'nothing', and 'yes' at least to 'something'. [133] Kicks are different. We can sanctionably say, "Well I had a lot of kicks [in the past]." The fact that we're having none now, or few, doesn't seem relevant … if you do something that counts in the society; an invention, a discovery, a contribution, then that can follow you. At any future time, should a person who's done such a thing try to make the conclusion "I'm nothing", it can be countered with "You've done this." Some of the assessable values are so structured that they are only relevant at a certain stage of life. For example, When persons 25 years old say in assessing themselves that they're unmarried, they're told, "No, you can't say that yet." These things are standardized; it's a matter of certain formal properties, that your age has to be X before Y counts as 'nothing'. (See also obligated others)

decomposable (An means of analyzing social norms based on inference separating the whole sense of the meaning of social action into its component parts) [36] I'm going to show some of the ways that I've been developing [an analysis of conversations] There will be series of ways fitted to each other, as though one were constructing a multi-dimensional jigsaw puzzle. One or another piece can be isolated and studied, and also the various pieces can be studied as to how they fit together. (However it should be emphasised that ‘decomposing’ is not something that only sociologists do, since ordinary members of society also decompose the talk of others to infer social meanings according to a set of norms they learn from socialization). [160] What members see is decomposable by them. That Members can decompose some event, situation, complex, whatever you want to call it, is no surprise at all, given the sorts of things we've been considering. That is to say, we've been talking about activities as being 'assembled'. And if that's so, and if Members can see that that's so, then, that they can take them apart would not be especially surprising. The question then is, having taken something apart, how do they put it back together again so as to find what it is that's strange. The way they seem to do this involves treating something that they see as a combination of parts, some of which have names. And to those nameable parts are affiliated standardized procedures for producing those objects in some combination. I'll give you an example… [161] a bum, is usually unshaven, wears very tattered clothes, and has a big flashy car. He's driving down the Massachusetts Turnpike and there's a police car sitting in the grass that divides the highway. He passes the police car, drives along below the speed limit, the police car starts to follow him, follows him along for a mile, finally pulls him over. The policeman asks for his license, the guy shows him his license. And now we get the following interchange. "What do you do?" "I go to school." "Where do you go to school?" "Harvard." "That's very nice. Where are you going?" "I'm going home." "Whose car is it?" "My mother's." Then the policeman says "Look, would you do us a favour?" "Sure. What do you want." "Would you get a shave? if you don't, every policeman on the Turnpike is going to have to stop you." That is to say, for such a person, how it is that they come to have gotten that car is a problem. And the incongruity is seen in that way; by virtue of the car being a possession, how would it come to have been possessed by such a one? [68] (Sacks offers another example) My parents live in an 'exclusive' suburb. And when I was a kid in high school I always used to walk around at night in the streets. And when you walk in the streets at night in exclusive suburbs, you're liable to get–as I was routinely–picked up by the cops. "What are you doing?" "Just walking." Then they would take me and stand me in front of the police car with the light shining in my face and call up the police station to find out if I indeed did [69] live there. This happened night after night. Finally, someone gave me the solution. If you bought a dog, that was the end. You never got stopped, and that has now become a matter of common knowledge. [163] (Student observation). It's Goffman's argument, the notion that persons feel that if somebody is acting in a particular fashion which they may doubt as being an example of such-and-such, then they're entitled to inspect him for other features that they assign to a such-and-such. (Sacks’s reply) Yeah but the fact that they did it is not sufficient. The first thing that we find is that events are decomposable. And the question is, "Could somebody have done that?" and it could be decided that they couldn't have done it properly, and now we get all these incongruity observations. Now the fact is, Goffman talks about incongruity but he does not tell us what [164] incongruity is. That's what I think I'm beginning to see here in this stuff. How it is that one sees it. He has not analyzed how it is that you do 'an incongruity', what makes it an incongruity. And I think I have the beginning of how it is that you do it.

dependence [143] Callers [to suicide help lines] start off with 'requests for information' which involve, then, that over the course of the conversation help will be offered without being 'asked for'. Because asking for help is something that one doesn't do with respect to strangers. So, "I need help" and "How does this organization work?" …

A: This is Mr. Smith. May I help you?

B: Well, I don't know. My brother suggested that I call you.

I want to stop here and show a way to analyze classes of statements for sorts of things we can find in them. I had these exchanges, where a person calls and says "My brother suggested that I call you" and "A friend of mine told me to call." I thought gee, that's curious. And I tried to think of a paradigm line, of which 'X told me to call' is an instance. It looked like the kind of line about which a psychiatrist would say, "This is a dependent person." And then I tried to see what it was that made it look like a matter of "dependence". I thought of the following. A kid comes into a grocery store and says "My mother told me to buy a dozen eggs." It looks exactly alike. It looks very much alike, anyway. And that's one of the major prototypic types of children’s accounts; naming some adult who told them to do some activity [144] One can then watch the development of an 'adult' by the shift in the use of such blank forms as 'I'm doing X (action) by virtue of Y (competent agent).' One may begin to see over the course of a child's growth that the child will be using himself as a competent agent for an expanding class of activities. On the other hand, where someone is a presumptive adult, about whom it's said, "They're infantile", we might find that they are producing a statement in which X is something that this person could do on their own, but the Y proposed is some other agent. (See also accountable actions; insiders and outsiders; obligated others)

entitlement [134] There seems to be a notion of 'entitlement', where by one examines some problematic outcome by reference to the *Member's theory of how that outcome is arrived at. [T]here are some very specially interesting things going on in this statement, "... what man wants a neurotic, childless, forty year old woman? No man." What has to be seen about that formulation, "No man", is that it's to be understood as 'no man Member'. That is, we're always talking about *classes and class members. In general, the notion that you can't get there from here has to do with [135] the ways in which persons arrange membership classes to do things; to find pairs, to get appointed to jobs, to become successes, etcetera. Such arrangements are taken as ways of assessing whether it's possible that you could do that, [this] works not simply to argue that you haven't any prospects, but also as a way of deciding what it is that's happening to you. For any *M.I.R. device category, there's a set of inferences attached to it, which are "common knowledge." Some person is a nominal member of a category, but feels that the set of inferences that are properly made about that category are not properly made about them. They can count down along that list of inferences and find "I'm not this, I'm not that," by reference of course to a rule of relevance as to what one is or is not. (See also categories)

eraseability [118] some things are in a sense eraseable; because they can be forgiven and some things are not. For example pneumonia like many [physical] diseases are eraseable. The cure removes it from the record. And the next occurrence is not seen as explainable by the earlier one. But suicide is not that way. If you make an attempt at some point in your life, then the fact that, for example, you went through therapy, or you were seven years old and now you've grown up, does not remove it from the record. That thing remains alive. So that if you kill yourself forty years later, persons will refer back: "See? He made a series of attempts." It's not an eraseable matter. (See also cumulative values)

ethnography [72] the trouble with [anthropologists’s] work is that they're… asking questions of their subjects. That means that they're studying the categories that Members use, to be sure, except at this point they're not investigating their categories by attempting to find them in the activities in which they're employed.

forms of address [36] there seems to be a fit between what the first person who speaks uses as their greeting, and what the person who is given that greeting returns. in choosing their form of address they can thereby choose the form of address the other uses. By 'form' I mean in part that the exchanges occur as 'units'. They come in pairs. Saying "This is Mr. Smith may I help you" thereby provides a 'slot' to the other wherein they properly would answer "Yes, this is Mr. Brown." (See also accountable action)

glances are actions [151] I'll give you an example of what I mean [by glances are actions], just from a lay approach to the matter. I was walking down the hall the other day, to give an exam to one girl. She was standing, leaning up against the wall. In between us walked another girl. She passed this girl first, and then me. And the girl who was [152] standing leaning against the wall looked at me and gave a shrug of her shoulders with a big smile, which I returned. And I don't think it was a big puzzle over what was going on. The girl who walked by was smoking a pipe. Now, the two of us knew what we were noticing. But that can be problematic. For example, on the Berkeley campus or in places in Berkeley, you often find interracial couples wandering around, one of whom is Negro, one of whom is white. And people who look like tourists, visitors to the campus, etcetera; that is, strangers, will stop to look at these couples, and then check out with others around. The question-form might be seen as something like "Am I in Rome, or am I here?" That is, Rome, genetically. "Do I know where I am so that I know that that's something okay or something odd." And when they do that, people will not infrequently just look back at them and give them a negative stare. As if to say, "Who the hell are you." And that's treated as very disturbing. The problem that I am stuck on at this point, and I don't have anything like a solution to it, is how does a glance become an action? What kind of a world do you have to build to make a glance an action? Let me start off by reading you a quote from an extremely important book, the title of which I also forget. I think it's called Sight and Sense, but you can easily find it given that the author's name, Von Senden, On pages 61-62 we get a report by duFoe of "a girl who only discovered at the age of twelve that she differs from other people in lacking a sense [a sight], and who now seeks to discover the nature of this unknown sense." And she says:

I posed myself a host of questions about this new and unknown state which had been described to me, and did my best to come to terms with them. In order to satisfy my doubts I had the idea of trying a strange experiment [156] One morning I again put on the dress which I had not worn for some time because I had been growing so rapidly then from month to month, and thus attired I suddenly showed myself at the door of the entry room in which my governess was already working at the window. I stood listening. "Good heavens Lucy," she said, "why have you put on that old dress that only reaches to your knees?" I merely uttered a few idle words and withdrew. This was enough to convince me that, without laying a hand on me, Martha had immediately been able to recognize that I had again put on the dress that was too short. So this was seeing. I gradually recounted in my memory a multitude of things which must have been daily seen in the same fashion by the people about me, and which could not have been known to them in any other way. I do not in the least understand how this happened, but I was at last persuaded. And this led gradually to a complete transformation of my ideas. I admitted to myself that there was in fact a highly important difference of organization between myself and other people. Whereas I could make contact with them by touching and hearing, they were bound to me through an unknown sense, which entirely surrounded me, even from a distance, followed me about, penetrated through me, somehow had me in its power from morning to night. What a strange power this was, to which I was subjected against my will, without for my part being able to exercise it over anyone at all. It made me shy and uneasy, to begin with. I felt envious about it. It seemed to raise an impenetrable screen between society and myself. I felt unwillingly compelled to regard myself as an exceptional being that had, as it were, to hide itself in order to live.

Most of Von Senden's work concerns the perception of patterns. If persons are blind from birth and they get their sight as adults, it turns out that certain patterns just can't ever be learned. They may not ever learn how to see a rectangle. They know what a rectangle is, and they can find it by counting the points. But where one sees, quote, at a glance… But what's important in the first place is to try to determine, in part, what it is that is seen in this fashion. And that's why it's sort of a pain to intellectualize this stuff such that you already talk about it as though, "I see a blob, and then I infer that it's my mother because she's a blob like that," when what you see is not that. And it's an extraordinary experience when it turns out that you do see somebody in something like that fashion. I just recalled, in Fitzgerald's novel The Last Tycoon, he reports a scene where this girl remarks that [157] walking in New York, seeing a man approaching, finding a whole series of properties of him which she doesn't especially like, it turned out it's her father…So where are you going to start to try to build a way of dealing with this stuff? We start out with the fact that glances are actions. That's the first fact. There's a beautiful report to this effect. A guy is looking at a girl, looks around to find somebody to exchange glances with, catches the eyes of another girl who looks like this one in some way, and turns away quickly. That is, he sees her sanctioning his looking. And how do we start to provide for glances as actions? I take it we have to start building *classes… the class permits you to see what it is that's there. It permits you to see. To liven this matter up, I'll read you something from The City of Plains by Proust. And Proust is an incredible sociologist, as you may know if you've read [him] - and if not you certainly ought to, even if you're not interested in literature. There is a scene where Proust is watching events take place in a courtyard below. He sees a whole sexual confrontation between two guys, which he describes absolutely fabulously. Then he writes:

From the beginning of this scene an evolution in my unsealed eyes had occurred in M. Charleaux. As complete, as immediate as if he had been touched by a magician's wand. Until then, because I had not understood, I had not seen. The vice - we use the word for convenience only - the vice of each of us accompanies him through life after the manner of the familiar genius who was invisible to men as long as they were unaware of its presence. Our goodness, our meanness, our name, our social relations do not disclose themselves to the eye. We carry them hidden within us. Even Ulysses did not at once recognize Athena. But the Gods are immediately perceptible to one another, as quickly as like to like. And so, too, had M. de Charlieux been to Chiupien.

It is the explanation that opens our eyes. So, the classes and their categories permit you to see. That's a start. It's not enough to make a glance an action. As some of you noticed and reported, it's not merely that some observer is seeing by reference to some category, but that the one being observed sees what the observer is, and is seeing. And we get into that whole jumble there: A, seeing that B is looking at A, sees what B is, and what B sees, etcetera, etcetera. A can see what B is, and what B is seeing. How is it that, given there's a whole bunch of classes available, A can see that? I have to introduce a notion that will help this problem along. Let's talk about there being, for some collection of classes, an 'order of relevancy' with respect to categories. [159] for now I'll only deal with classes that have two subclasses. For some classes, A, being observed, sees that B, doing the observation, sees that A is a member of some contrast class such that A is one part of a paired class and B is the other. Where for each, there is an order of relevance which provides for the other as being an observable, and… the order of relevance of each was available to the other. With that, we can begin to handle some rather nice kinds of events; you can begin to see what the following kind of trouble is. A girl looks at a guy. And what he does is, he takes her glance and then looks over his shoulder. And what's going on is that he knows enough to know the sort of thing she's looking at, by reference to the class - she's looking at 'an attractive male' - and he has that sort of insecurity which provides that he's not eligible, so it must be somebody else. This complementarity of orders of relevance permits us to see how persons are operating when they talk of themselves as 'worthless' or as 'nobody'. And it's this complementarity - that one knows what the other classes do - that begins to tell us what may be going on, in part, where some guy is looking at a girl, now looks around and catches the eye of another girl who is somewhat similar to the first, and turns away. That is, he does not seem to just continue looking around as though, "Well, she's not a male so how would she know what I was doing?" Or, for example, some of you report a person will be walking, sitting studying, just doing nothing, and then see somebody looking somewhere, and quote, follow the glance, knowing that there would have to be something at the end of the glance which is worth looking at. And if you can see what it is that is doing that looking, you could have a pretty good idea of what it is that would be at the end of it. So this complementarity is equally as crucial as the fact that one is able to see what somebody with whom you are a member of a class in common is seeing when they look at you, or another. [165] So the norms don't seem to be in doubt (when judging the legitimacy or illegitimacy (*averageness) of an action); that is, the norm that provided for the incongruity that provided for the noticing is not something that seems to be held in doubt. And the notion that it's 'reinforcement' that you get is somewhat obscure. Though the question certainly has to be asked, what is it that people are doing when they exchange these glances? We've already observed that they get exchanged with whomsoever; there's no special credentials involved, except class membership. Another, equally extraordinary part of this - and why, again, 'reinforcement' is a tremendous gloss or oversimplification - is that one takes it that they know what it is that you saw, that they saw the same thing, and they know what you're smiling about, and they make the same assessment. Their smiles tell you that. Now, that that gets done would seem to involve a fantastic kind of social integration. (See also inference making machine; order of relevancy)

greetings [170] There are several things we want to notice about greetings, apart from the fact that they are *adequate complete utterances. Greetings are paired. And by that I mean simply that if A picks a member of one of those things, then a proper move for B when he has an opportunity to speak - right after it - is to pick a member also; the same, or another. So one party's use of a greeting provides for that minimal exchange, "Hello," "Hello."… [171] We need to distinguish between a 'greeting item' and a 'greeting place'. Where, then, something is a 'greeting' only if it's a 'greeting item' occurring in a 'greeting place'. If a greeting item occurs elsewhere it's not a greeting, and if some other item occurs in a greeting place it's not a greeting -though some items that are close to greetings might take on the character of a greeting by occurring in a greeting place. We need, then, to be able to say that there's a 'greeting place', and that any 'conversation' has it. And I take it we can say that there is a greeting place in any conversation… here is no rule of exclusion for greetings. People can know each other 35 years, talk to each other every day, and nonetheless greet each other when they begin a conversation. But take other items, for example 'introductions' … it can be said that there are rules for their historical use. At some point in the history of persons' conversations, introductions are no longer relevant. [172] we can distinguish between the greeting item in the greeting place, and the greeting item elsewhere. That is, somebody can say "Hello" in the middle of a telephone conversation, where what they're doing is not 'greeting', but checking out whether the other person is still on the phone, and a variety of other things like that. Thus, the greeting item, to be 'a greeting', has to occur somewhere in particular, and we can say, then, that there's 'a place'. And in that greetings are relevant for any conversation, that place in which they are recognizably not something else, but 'greetings', is present for any conversation, whether there is a greeting item in it or not. (See also A-B reduplicated)

‘I don’t know’ [80] when you see "I don't know" appended to some statement, that's what it seems to be doing - providing that "I'm not entitled to say this", that is to say, "I can't defend it professionally", if it's a matter of professional information.

ideology [136] I take it that *'imitation' is one of the basic categories one wants to focus on when one talks about the phenomenon of ideology. Because it makes noticeable that there are a whole range of things that [unentitled] persons obviously can do, which are by and large not seen as things they [should be doing] the sense of things that a Member can do, and the addition of capacities is treated as more things that can be imitated. And that's an extremely interesting kind of blindness, if you want to put it that way. It shows you. [137] the power of this procedure because it's apparently a perfectly consistent and reasonable way to talk, and the materials are thus never shifted over to be seen as, "They can do those things." In rather more abstract terms, we can come to see a way in which such categories as *'imitation' and '*phoney' provide us with something very central, in that they serve as boundary *categories around the term 'Member'. (See also entitlement)

imitation [135] I'm not using the term with respect to actors in stage performances, but where persons observe some activity and say, "He's imitating." I came across an extraordinarily interesting use of this category in some of the older ethnographies, dealing with the situation of Negroes in the pre- and post-Civil War periods in the South. Again and again I found references to the activities of Negroes as 'imitating whites'. And they were characterized as being 'marvellous imitators'. [136] Now, we want to ask what does 'imitation' consist of, procedurally? How is it that some behaviour is seen as 'imitation'? One of the central things that seems to be involved is this: When one normally deals with the activities of a Member, apparently one takes it that they have some right to do some class of activities, and that when one engages in making out what they're doing, one takes it that what one sees them doing is what they are doing. 'Imitation' seems to involve a way of characterizing some action which somebody does when they are unentitled to do that class of action… So 'imitation' becomes a category which involves the construction of a parallel set of knowledge for those unentitled Members, where it doesn't happen that as they do something one finds that there is 'the doing', but as they do something one finds that they're able to imitate. It's noticeable in relation to this, that if the capacities of some persons are treated in this way, then one finds that certain sorts of accounts that can be applied to Members in general cannot be applied to them. So one finds that they can't be found to be 'responsible' for what they do, in a non-trivial sense. If you watch, let's say, the way that children's suicides are described, you'll see that it's not enough to say that the child was depressed, the child got a gun and shot itself or whatever, but there's an added item: How it is that the child was free at that time to do it; for example, their mother went out of the room and then the child took the aspirins. That is, part of the causal account is some competent person's actions which permitted this thing to take place (See also dependent; entitlement)

inference-making machine [200] [An inference-making machine can] deal with and categorize and make statements about an event it has not seen. i.e. uncover an explanation for an event, or vice versa, an event for an explanation [199] Now let's try to begin to consider what our task is if we're going to build [an inference-making machine]? The first rule is do not let your notion of what could conceivably happen decide for you what must have happened. (For example it is not the case that if somebody does something pretty simply, pretty quickly, or pretty routinely, then [the explanation is likewise going to be simple routine or quick). The second role is the sort of events it can handle are sequential events. For example, if we have (a), (b) ... (d), then the machine’s job is find out what (c) is. What we have is roughly something like this: A machine user let’s call him ‘A’ examines a scene where the police have been called to ‘B’s’ house. He knows that the scene is 'a family problem'. (So (a) is the family quarrel, B has told A that his wife "stepped between me and B’s child, and B "went to move her out of the way." . And then event (d) is her sister had called the police. Therefore (c) must be the grounds for the police to have been called). That is, apparently on some piece of information the police have come, and that piece of information is the thing that A has guessed at. [Because] A apparently knows, what good grounds [there] are for the police to be called to a scene. And he's able to use those good grounds, first to make a guess, and then to assess the correctness of the answer to that guess. A knows [all of this] essentially only [through the] set of terms that B uses to name [the events]; that is, there is something called a sister, something called a wife, something called a child. It seems to me that the information that is being used by A is held in terms of collections of these categories… To get a sense of the way in which the inferences that can be made from a story are geared to these categories, we could try, for example, using different categories. What if it were, not "her sister", but 'a neighbor' who had called the police? A possible inference in that case would be that the grounds for calling the police had something to do with 'creating a disturbance'. Or, for example, just shuffle the one category around a little bit. Or, for example, just shuffle the one category around a little bit. Would the same inference be made if it was 'my' sister, not "her" sister who had called the police? The rules with respect to who owes what to whom, and who takes care of whom may be so formulated that those things matter a great deal. The inference in this case might then be, not that the husband had produced some activity which served as good grounds for calling the police, but that the wife had done so. And that is extremely important because it is an awesome machine if one needs to know only that it is "my wife" and "her sister". And you can [202] do this because that holds for every like unit in society, such that you don't need to ask for example. (you substitute the actual individual mentioned for the *category or *class of individuals)…. It is not merely that the notion that you need to know a great deal about somebody before you can say this or that about them may be a lot of nonsense, but the way that society goes about building people makes a nonsense of such a notion. That is: A task of socialization is to produce somebody who so behaves that those categories are enough to know something about [them]. [203] [T]he fact that some procedure which has a correct way of getting done, gets done correctly… may be quite crucial in permitting persons to find the sense of an event which happened by reference to those procedures. [furthermore]… a person who stands in the position of having some procedure which has a correct basis for use applied to them, stands in the position of having that procedure presumptively correctly applied. This is what Sacks calls *Job's Problem … A central dilemma for users of the inference making machine is that some procedure which has a proper way of operating, may not so operate … A problem that people in a range of circumstances can be faced with is: Is there a way available to provide for some event to have happened, apart from the normal and proper way these things happen? Job's problem focuses on a central problem in the way [204] that persons go about orienting to the occurrence of events, and that is, that it is somehow extremely important that the inferences they do make can be taken as correct, and thereby that those persons who produce those activities which are described by these sequences so behave as to provide for the fact that these sequences do describe them … A problem for a sociology interested in describing socialization will consist in large part of how it is that a human gets built who will produce his activities such that they're graspable in this way. That is to say, how it is that he'll behave such that these sequencing machines can be used to find out what he's up to. Under that notion, then, we would propose that at least one core focus for trouble would be persons who are so socialized that they don't permit these sequencing machines to be used on them. And that is one way that 'psychopathic personalities' behave. the psychopathic personality is [205] reported to be that person who, at any given point in their behaviour, you never know what's going to happen next. You're never able to say "Here is an Nth point in this sequence, and now X, Y, and Z will come." [209] Let me mention one further thing, which is again relevant to the conversation we're examining. When we think about… scientific facts, we tend to pose problems in the following way: If it's the case that something has occurred, then our problem is to explain it. Now, with such things as lies, untruths, confabulations… we've got to notice that something like a reverse procedure is very much used… The reverse procedure consists of the following. In deciding among possible competing facts, one may decide that that fact occurred which has an explanation, and that fact that hasn't an explanation did not occur…[210] one can choose among facts according to the presence or absence of an explanation. It's absolutely routinely used... I'm only proposing that that's the way it's done. At least in this society, facts and explanations have more than a one-way relation to each other. That is, it isn't the case that if something has occurred, that sets the problem 'construct an explanation', but the notion that persons hold of possible facts is that those facts are possible for which there is an explanation… One can't merely say "Well, I saw it. You explain it." Something proposed to have occurred can be treated as not so, by virtue of the fact that there's not an explanation…That's important in this society, given the fact that Miracles are no longer usable. And Miracles are that class, in part. They're events for which there is no account, which now systematically would be given an account; i.e., an account not of this world. But most persons who consider themselves to be modern individuals don't [accept that explanation] [211] That's a classical problem. It can always be raised if you have a limited amount of research, where, then, if you propose there's a phenomenon, it is said that you have nothing but the working of a random distribution, which you happen to have caught at some point … And it's raised more than occasionally about psychic research, and I suppose more than occasionally it turns out to be relevant … Or, for example, people say "How does it happen that there have been two earthquakes on this day?" and then they go about constructing some explanation: The gods are angry.. then you're liable to find something which looks like order. (See also order of depth; order of relevancy; subversion)

'insiders' and 'outsiders' [140] two classes of 'others', and said something to the effect that one class were persons with respect to whom there was such a bond of obligation that you could directly say to them "I need help"-, for example, 'family'. We can talk of the two classes as 'insiders' and 'outsiders'. (See also accountable actions; dependence; obligated others)

invitations [141] the proper way that activities get started. Intimacy between the persons can be measured by the fact that don't use invitations. (See also ceremonials; questions)

Job's problem [203] a person who stands in the position of having some procedure which has a correct basis for use applied to them, stands in the position of having that procedure presumptively correctly applied … Job, the rich, good man, had lost all his wealth, his children, all his possessions. His friends come to him, and there are series of long discussions. What his friends propose is, look, you take it that God punishes the wicked and rewards the good. We take it the same way. Your situation is understandable only if you're guilty. So confess. And for Job the question is, "I don't know that I'm guilty. I'm convinced that I'm not guilty. But then how could this have happened to me?" [61] he is not about to acknowledge that. But most people do. Most people, when they get into a situation, will say, "What did I do wrong?" or "What did I do to deserve this?" That is to say, treatments are 'proper treatments'. And one isn't in a position of saying right off, "You're treating me wrong." Rather, one finds, the treatment occurred and it must be about my action. [203] Job's problem focuses on a central problem in the way [204] that persons go about orienting to the occurrence of events, and that is, that it is somehow extremely important that the inferences they do make can be taken as correct, and thereby that those persons who produce those activities which are described by these sequences so behave as to provide for the fact that these sequences do describe them. (See also Subversion (Adam’s problem); inference-making machine)

jokes [175] The first thing that's important about jokes is that to use one is something like buying a drink among a bunch of people: They come in rounds. And if some person tells a joke then every other person present has the right to tell a joke. [176] … when jokes are told they're things that are 'going around'; they're quotes. So they're unaffiliated remarks, and in that sense it's hard to say about somebody that the fact that they told some particular joke has some special significance. They just heard it, and now they're repeating it… [Their function is that] persons can monitor the conversation, watching either for silence or for the approach of something dangerous, and start a block of talk by flicking in a joke, thereby giving each other person their chance to talk, and to talk 'safely'. [104] [A] problem for a questioner at a press conference is that the answer might be nothing more than a "Yes" or "No". Or, for example, the Kennedy format, responding to a question with a quip. And that in its way is akin to "This is Mr. Smith may I help you?" as a way of getting the other's name without saying "What is your name?" and also akin to laughing when someone says "I'm going to kill myself as a way of [105] refusing to give help without saying "I won't help you." The Kennedy quip is a way of refusing to answer a question without saying "I'm not going to answer that question." Because after the quip, the fellow who asked the question can't come back. And whether anybody else will come back to it remains to be seen. (See also ceremonials)

knowledge (status of) (Compare opinion) knowledge acquisition (See category)

legitimacy and illegitimacy [161] In any given circumstance the person who says that is correct or not, they are in fact employing a procedure which might be employed if they happened to be together with someone … That is to say, one thing which seems to be observable as an 'incongruity' involves two persons being present, a man and a woman, where there's some special difference between them: A worn old man and a very young pretty girl … And now they get noticed, an exchange of glances takes place, and the question that seems to be asked is "How did he get her?" ... What one is doing is employing the procedure by which persons properly come together and finding that that does not produce these two persons as a pair. And one can then produce an explanation… So that the procedures whereby persons come to be in some [162] combination, or come to have some object, seem to describe in part what it is that you've observed, and what I had… never really been able to figure out - how it is one sees an incongruity, and also sees the possible illegitimacy of some combination. Now, in some of the reports what we seem further to find is that where persons are concerned, units of larger-than-a-member are observable. People can see 'a family', for example. That may be trivially obvious, but it's very important, and it has to be achieved in some way. One of the reports has a powerful instance of the relevance of persons seeing such a thing as a 'family'. One of you reports that you were driving along and a car pulls up and stops. It doesn't seem to stop anywhere special, just pulls up and stops on a street. The door opens and a girl of about 18 charges out, runs across a lawn and stops, and starts shrieking. In the front seat are an older man and woman. The guy jumps out of the car, charges across the lawn, comes up to the girl and gives her a smack right in the face. At which point some of the passing cars slam on their brakes, and some people start getting out of their cars. The man and the girl stand there, face to face, screaming at each other, and then he just grabs her and drags her back to the car. And people look at each other, shrug, and say "Oh well," get back in their cars and go on their way, taking it that it's not after all a kidnapping scene or an attempted murder, but it's 'a man and his daughter'… (See also average; uncertainty)

measurings [113] A variety of items on which persons can monitor their own states. Either routinely, or, when asked. [114] They have the information which permits them to engage in some consideration and then give an answer as to a current state or current variation from a prior state. The items are things like sleeping, appetite, etcetera. Measurings are variant categories [deviations from normal] which are also standardized, like 'poor' and 'great'. There are whole bunches of terms: Fantastic, cool, terrible, whatever you want. It's essentially the same. We'll call them 'directional differences'. Minus and plus. So we get 'normal', 'minus', 'plus'. Variations from 'normal' are noticeable phenomena. They're noticeable by reference to whatever it is that's "normal for me". And it's the fact of the variation which is relevant to some state being noticeable, and not what the normal state's features are. These variations are individual dependent you don’t have to have an equivalence across groups of individuals for them to be valid. It's also to be observed.. that the variation categories are standardized, and they're standardized without respect to what they contain or what the normal contains. Any Member can employ the set of categories to formulate their current state; that is, they can say 'normal', 'poor', or 'great' without reference to what it is that that stands for, or how what it stands for compares with what anybody else has theirs stand for, and they can talk about it. [115] These are clearly measuring categories of a sort. And one might have the idea that, given the way measuring is done, these are "rough versions" of something that is [mathematical]; that persons are talking loosely about something which our business would be, in studying it, to find out what the specific measures are. What I'm saying is, there aren't any such measures built into the use of these categories. I'm pointing to the fact that here's a medical device of a sort. Its power comes from its emptiness.

Members (Sacks uses the term ‘member’ in two ways: firstly, to designate anybody can who make assumptions about another’s behaviour because they belong to the same class of individuals within their society, i.e. white people or women. Secondly, the term members can designate anybody who can be turned to in order to checkout another’s behaviour as normal because they belong to the same class of individuals within that society. Members are therefore the ’us’ as in ‘us and them’. Although how membership is defined is very much contingent on the interlocutors's definition of the situation. Someone who may in other circumstances would normally be defined as an outsider may in certain circumstances be defined as a member in others: for example a person who has not grown up in the UK but is nevertheless a UK citizen would be one of ’us’ when talking about illegal immigrants.) [164] In the first place, if one were engaging in some device for getting reinforcement, (or for ascertaining whether their judgment of *legitimacy or illegitimacy for an action was accurate of not. One might seek it from someone else). It might be supposed that one would want to know the status of the person one uses to get reinforcement from … You might ask… a priest, or ask your parents, or ask somebody who has special rights. In any case, it might be presumed that you'd be concerned to check with somebody who you knew to have some information about it. But … apparently what we find is that it's pretty much an 'anybody' who can be turned to for that checkout; 'anybody' as long as they're a member of an appropriate *class, so let's say, a woman turns to a woman, or a man turns to a man. You don't ask for their credentials in the first place, and if they return your glance, give you the smile back that you give to them, or the disapproving glance, that seems to be okay. And that's a fantastic kind of simplicity. (See also categories; glances as actions; inference-making machine)

M.I.R. category device (Membership, Inference-rich, and Representative category device) [85] MIR category devices are the central machinery of social organization. I've very frequently found, as anyone can easily find, that especially in the early parts of these conversations certain questions are prominent; questions like "What do you do?" "Where are you from?" etcetera. I wanted to see if there was some simple way that I could describe the items that those questions contain, so as to provide for their occurrence by rather abstract descriptions. It seems that there is a *class of *category sets [90] we can call 'inference rich'. By that I mean, a great deal of the knowledge that members of a society have about the society is stored in terms of [them] the inference-rich character of these categories constitutes [a] warrant for their occurrence in early parts of first conversations: When you get some category as an answer to a 'which'-type question, you can feel that you know a great deal about the person, and can readily formulate topics of conversation based on the knowledge stored in terms of that category... [A]ny member of any category is presumptively a representative of that category for the purpose of use of whatever knowledge is stored by reference to that category. So, for example, a foreigner comes to the United States and you find yourself asking them about the political situation in Ghana, or how they like the food in the United States, without reference to whether they stand as a member of the Gourmet Club of France, or don't ever eat out, or aren't interested much in food, or are just ordinary citizens, so to speak. But one finds that it's done. And it's done for any of these category sets [91] [H]ow [is] a class of social control devices… set up and is used [?] And: through interrogation and self reports For example, the hours between the assassination of President Kennedy and the determination of who it was, and thus what category it was that performed the act. If you have access to a variety of materials from that time, you can see persons reporting themselves going through "Was it one of us right-wing Republicans?" "Was it one of us Negroes?" "Was it a Jew?", etcetera. These internal control devices all seem to be built on, and have their power by virtue of, this very simple apparatus the M.I.R. device, which is utterly disjunctive to whatever these groups happen to be, or whether they happen to be 'groups' in the organizational sense of the term. [93] These control devices are unlike *scapegoating.

M.I.R. category device modifiers [94] MIR category devices presume that for some category there is a set of things known about a member which can be applied to any member, for example that being on the girls' tumbling team is presumably something very gauche, and that someone who is 48 is past their prime, which any person now talking about such a category membership has to come to terms with. MIR modifiers consist of attempts to provide that what it is that may be said about any member is not to be said about the member at hand. "I'm 48 but I look and feel younger" "She's on the girl's tumbling team but..." For a further example, take the following quotation:

[97] A: What interests did you have…?

B: I was a hair stylist at one time, I did some fashions now and then, things like that…naturally. You probably suspect, as far as the hair stylist and, uh, either one way or the other, they're straight or homosexual, something like that

In this case, while it might not be proper for this man to say about himself that he's troubled by possible homosexual tendencies, he finds a way to invoke a subset of occupational categories, "hair stylist... fashions ... and things like that", which constitutes an adequate basis for inferring homosexuality. And in his subsequent talk he proposes that such an inference has "probably" been made by the other. Apparently, then, there are ways of introducing a piece of information and testing out whether it will be acceptable, which don't involve saying it… So they're the basic system of incentive for persons to do a variety of things… so all the various occupations are engaged in trying to sharpen up their images so as to make it attractive for persons to come into them….I have by and large been talking about negative information stored in these categories, but they obviously provide that system of rewards which any young person can expect by virtue of becoming a member of any category that they can become a member of … since to become a member is to make stateable about yourself any of the things that are stateable about a member. (See also perspective. Compare tautological statements (as anti MIR devices))

microscopic sociology [72] There are … matters of a deeper sort which are perhaps relevant to why sociology took the course it did, and they're intrinsic in Durkheim's work. One of them is the notion that the order of social events is macroscopic, in the sense that you had to assemble lots of events to find statistically what it was that was doing the work. I think one can begin to see, in the stuff I've been talking about, that it may well be that things are very closely ordered. And what we have may be something like the following. There may be collections of social objects - like "How are you feeling?" -which persons assemble to do' their activities. And how they assemble those activities is describable with respect to any one of them they happen to do. That's a different kind of order to a social world. [122] I've been interested in the possibility of orderliness in what people do for a long time, and not known that they could be working in that kind of order. Now, the way I work has been called 'microscopic' with, then, the usual sociology as 'macroscopic'. And it's not a bad distinction. But then it's proposed that social events are not closely enough ordered so that we can get results at the 'microscopic' level of investigation. I take it that we just don't know whether or not that is so. Certainly there has been an argument, and certainly the statistical position has won out. Durkheim posed the matter - which is the basis for the statistical approach to sociology - that if you take the statistical figures on suicides, you find quite an order. And you can study those. And construct theories. Then he says if, however, you deal with such things as the accounts that accompany each suicide, you don't find order at all. They're hastily made up, by low grade officials, etcetera. But he did not in fact attempt to deal with the accounts. And, in fact, his arguments as to why you shouldn't do non-statistical work were statistical arguments. And it may well be, for example, that the accounts of suicides are closely ordered phenomena. Another objection to the way I work is that it seems to be enormously laborious, and sociologists are not given to doing things slowly. But one reason I'm operating as I do is, I take it that the big problem is not that we know that social events are not closely ordered, but that we wouldn't know how to describe them to see whether they're closely ordered or not. I want to see if such work can be done in the first place. Then we can repose the issue of where are social events closely ordered and where are they not closely ordered. But it's then an empirical issue that has to be discovered. It couldn't be solved by an argument.

neurotic [146] [A neurotic is a person who] doesn't see rule violations which have had no consequences as involving *Class II rules, but as involving Class I rules with no time-bounds on their operation … Persons in such situations live out a great part of their lives under the sense of the impending consequences of their violations of Class II rules [which they see] as Class I rules. [147] That's a considerable part of the 'neurotic sense of guilt'; an ever-present sense of guilt which consists of their knowledge of the set of Class II type violations that they've done, like not loving their parents, etcetera, for which they haven't - yet - been punished. I think, a sense of what it is that the psychotherapist is talking about when he says that neurotic adults do not have a good sense of reality. And also when they say that they remain children, and that the projected operators in their presumptively adult lives are always parents.

normal [114] For each [*measuring] there is a category that *Members use, and that category is 'normal'. That is an extraordinary special category. Each Member employs it. Questions can be asked about it from one to another. But for a large variety of uses of the answer to questions within which 'normal' might be an answer, and for a large variety of the monitorings which might go on by oneself without anybody asking you anything about it, it's quite irrelevant what, for any given person, that notion 'normal' denotes. It's irrelevant whether it's similar to or different from the features that anybody else uses to decide that on some item they're 'normal'. So 'normal' is a standardized category, where whatever it refers to for any given person doesn't have to be specified to control its use. (See also average; class)

obligated others [56] There is [both a large and] a small group of persons included in the circle of persons who routinely use *ceremonials. Call the larger group 'others' and the smaller, a special class of others persons. [The second group are persons] who are causally bound are obliged to give help when help is asked for, for example, they may be kin, One of the ways [in which] they stand to each other is, if something happens to [the troubled person], then, whoever it is that might be trying to discover why that thing happened, could refer for explanations to these others [as being responsible]. [In that sense they are] 'causally bound' to the person who may have trouble. And that could quite easily make it apparent how it is that if such a one is turned to for help, they have a feeling of obligation. [T]he question would be asked, well what was up with that family that [person] should have killed herself? Further, somebody who is not a part of this small group of others can become causally involved by virtue of the fact that [person] has asked them for help in some way and been turned down. If something then happens to [that person], it seems that even if you aren't one of that small group of others, you know about the fact that [person]was troubled, how come you didn't do anything? So knowledge of the trouble is often sufficient to bring one into causal involvement. (See also ceremonials)

opinion [262] The notion of 'opinion' as contrasted to knowledge (and Plato made a great deal of the difference between them) and the sheer introduction of a notion of 'opinion', provides in part for professionals' talk to laymen. Because one of the characteristics of 'opinion' is that it's something which lay persons are entitled to have when they're not entitled to have knowledge - in the sense that they can offer it without ever proposing to have to then defend it … the fact of a distribution of knowledge which provides that professionals know and laymen don't know might seem tremendously interruptive unless you had some mediating device, like 'opinion', which would permit laymen to keep talking even when they find out that they don't know. Otherwise they might not have any way, for example, of even turning to a professional.

order of depth (a heuristic device that judges the extent to which an individual e.g. husband, wife, child, can be successfully substituted for the category ‘husband’, ‘wife’, ‘child’ without their idiosyncratic and individual characteristics distorting the information) [202] One of the things we always want to be watching for is to see how simple or how complex this animal is. In this regard I'll raise a question but not attempt to answer it here. There is what we can call an 'order of depth' in dealing with various kinds of occurrences.

order of relevancy (a heuristic device by or through which we can begin to locate by reference to circumstances certain salient facts from the multitude of less salient ones that obviously do occur in any complex social interaction). [158] It's not merely that some observer [‘A’] is seeing [that which is observed ‘B’] by reference to some category, but that the one being observed [‘B’] sees what the observer [‘A’] is, and is seeing. And we get into that whole jumble there: A, seeing that B is looking at A, sees what B is, and what B sees, etcetera, etcetera. How is it that, given there's a whole bunch of classes available, A can see that? I have to introduce a notion that will help this problem along. Let's talk about there being, for some collection of classes, an 'order of relevancy' with respect to categories. It seems that a set of circumstances can provide that order of relevancy for some membership class. If the circumstance is that A is being looked at by B, that in itself might inform the consideration of what [159] order of relevancy is operating here. And if there is an order of relevancies, we can begin to locate certain facts that obviously do occur. (See also glances as actions)

outsiders (see 'insiders' and 'outsiders')

pairs [54] The opening lines of conversations … seem to come in pairs. And that one person could choose the form of greeting he used. And that if one person could choose their own they could choose the other's. There is a symmetry in pairs. (See forms of address)

perspective [95] Some of the category sets of the *M.I.R. device have to be differentiated from others in some special ways. If one considers categories like age and social class, in contrast to those like race and sex, one finds some rather interesting differences. What you have with these latter sorts of categories is… if any *Member hears another categorize someone else or themselves on one of these items, then the way the Member hearing this decides what category is appropriate, is by themselves categorizing the categorizer according to the [96] same set of categories. So, if you hear B categorize C as 'old', then you would categorize B to decide how you would categorize C. And again, the same procedure works for such a thing as social class. This sort of operation is probably basic to something which sociologists are talking about as a generic matter … and which Members also use. And that's the notion of 'perspective'. (See also measurings; normal)

phoney [134] If what you think you are not *entitled to actually happens to you i.e. if somebody marries you, or you get a job or a promotion, there is a notion of ineligibility that can get used even though it seems that you're eligible by virtue of having been selected. And that's the notion of 'phoney'. "I'm a phoney" focuses on just that sort of thing. The most exquisite statement I've ever seen of that sort was written by a woman in the course of a psychosis which has been named 'depersonalization'. She says, "...the feeling itself is one of unworthiness, in the way that a counterfeit bill might feel when being examined by a banker with a good understanding and appreciation of real currency. (See also entitlement; imitation)

private calendars [84] one of the most exquisite kinds of things that young married persons do with each other is, they say things like, "Kennedy was assassinated two weeks after we got engaged." I want to give the name 'private calendars' to that sort of talk. I want to note that married couples, each one, by themselves, independently, construct these private calendars. And what private calendars do is to provide for the locating of, not only events within that relationship, but events of the world in general, by reference to the relationship. Further, these calendars are 'causally powerful'. What I mean by that is, there are all kinds of events which can be explained by reference to the relationship. There is a generic statement: 'Because A did X, B did Y', where one can substitute for A, 'wife' and for B, 'husband', and substitute for Y the event to be explained, and for X the activity which can explain Y. This provides a large class of sensible statements which persons in units like husband-wife are able to employ. Indeed for many events, such statements have to be employed; that is, for many events, such an explanation is the only sensible explanation. So it's often said that while you can give a whole list of explanations for why it is that somebody succeeded, in the last analysis it's because of his wife. It's said without knowing the guy, or knowing anything else. Another sense in which the private calendar is causally powerful can be seen in the paradigmatic statement, "That was before I met you and I was [85] lonely then." There is a class of logical statements which the logician Nelson Goodman named, and pointed to as creating very basic problems for the philosophy and logic of science. He calls them 'counterfactual conditionals', of which an example is, I think, "If one had lowered the temperature to such-and-such a degree, then the following would have happened".. Many scientific statements are made that way, and Goodman argues that there isn't currently a logic providing for them. But counterfactual conditionals are nonetheless routinely used, and they are, nonetheless, enormously powerful. Which suggests that perhaps a logic can be invented, or that they're building on something very strong... private calendars [are] ways of building up, in deep and repetitive ways, the relevance of 'you'. if we compare these private calendars to everybody's calendar, then there's one striking difference between the two of them. And that is, everybody's calendar has, and private calendars do not have, guaranteed continuity. Everybody's calendar runs on into the indefinite future, without regard to anybody in particular being present. Private calendars end when 'we' end. The end of a relationship, in one way or another, can provide that there's no more events on the private calendar. [86] We can see the widowed person saying, when they say "Nothing happens anymore", is that with regard to the private calendar whereby events between me and my spouse happen and the value of my life is found, no more events can occur on it… The task is at least programmatically simple, whether it's easy to do is another question. It involves bringing them back to the use of everybody’s calendar, whereon events can still occur sensibly in their lives. (See also category; inference-making machine)

proverbs [70] Things -- strictly traditional pieces of information --[which] have to be affirmed [by *Members] or membership is not seen as something both of us hold in common. [73] proverbs, may give an enormous understanding of the way humans do things and the kinds of objects they use to construct and order their affairs [119] what is *normal gets incorporated into things like proverbs and becomes very stable, odd events are just sloughed off. They don't get incorporated. The fact that it's odd is enough to mean that one doesn't have to consider it on this particular occasion. What you get is, "Those things happen, sure, but..." [189] [Proverbs are] pieces of knowledge that are organized atopically… atopical is not meant in the same way as 'abstract' [when defined as being] distinct from 'concrete' thinking… the proverbs themselves are quite abstract. 'A rolling stone gathers no moss' doesn't contain any reference to a particular rolling stone, a particular kind of moss, etcetera, etcetera… [190] proverbs can be seen to constitute a very clear example of whole collections of pieces of knowledge that are organized atopically. George Homan’s Social Behavior: Its Elementary Forms pp 1-2, proposes as a possible "shortcoming" in 'atopical organization'

What makes the subject of everyday social behaviour a chaos is that each of these maxims and proverbs, while telling an important part of the truth, never tells it all, and nobody tries to put them together ... but uses them ad hoc within the range of situations to which each applies, dropping them as soon as their immediate relevance is at an end, and never asking how they are related to one another.

But there are some obvious virtues to having a body of knowledge organized in an atopical fashion. You get a piece of knowledge, like 'a rolling stone gathers no moss', which is in the first instance correct about something. If you paraphrase it into some particular domain, like "a man who doesn't settle down doesn't gather possessions", then it may not have the same kind of correctness; it may be questionable … One of the most striking things about proverbs is that while on any occasion of use they may be, for example, inappropriate, people do not propose about them, "Is that so?" "What is your evidence for that?" If there is a question about them, it is in terms of, is it appropriate to apply that proverb to this person, activity, etcetera? That is, proverbs are in the first place correct. And that can be accomplished by formulating a proverb from a domain within which it is correct, and having it always be used elsewhere. In that way, instead of constantly revising a body of knowledge [191] by reference to the discovery that it's not correct here, now, for this, you maintain a stable body of knowledge and control the domain of its use. … they have the character of being potentially descriptive or relevant. Persons learn them and have them available for use. They don't… learn them on the occasion of their appropriate use …Both maxims and descriptions can be produced as proverbs, formidable as single sentences. "A woman's place is in the home" is an instance of the former, and for the latter, they are frequently those 'proverbial phrases', "stacking the deck", "hanging by a thread", "barking up the wrong tree", etcetera. Both types of knowledge, then, can be had and used via the single sentence, and thus it's not necessary to have some combination of sentences so as to minimally understand and transmit such information. Having that, you have a setup designed to permit you to learn new members of that *class of information much more quickly than might be otherwise possible … at that point when children can make single sentences, their task is now to see what it is that can be packed into that form they've learned… [192] Children learn that strings of words can assemble potential descriptions, only in certain arrangements. But they're [also] learning that there is a correct way of assembling potential descriptions, apart from the particular occasions of their use … These sorts of considerations may have a bearing on a classical controversy, mainly within linguistics, which concerns the question: Are [193] grammar and meaning separable? It may be that "Are they separable?" is not the problem, but that if 'meaning' is, in part, reference to something, and grammar is understandable apart from reference to anything - that is, formally correct in some way - then it's not simply a technological linguistic question, whether it's so that they're separable, but that it's an essential fact of language that they are separate. (See also A3N device; erasability; Job’s problem)

questions [177] I'd asked people in the class to write down the first lines of what they took to be 'pickups'. I got 60 first lines, of which just under 60 were questions. [101] A person who asks a question has a right to talk again afterwards. And that rule can provide a simple way of generating enormous masses of sequences of talk: Question, talk, question, talk, etcetera, etcetera. We can say it's a rule with a repeat device. But what else can we get from it? …[177] The conversation goes at least something like A-B-A. [178] Things like "Don't I know you from somewhere?" "Didn't I see you at such-and-such a place?" "Didn't you go to such-and-such a school?" "Aren't you So-and-So?", etcetera. All of these provide for the fact that it may be the case that we know each other, and if we do, then this conversation can take place as 'a further conversation' … very frequently the first question will be a request. And the request will be such a thing as can be asked when any two persons are physically available to each other. (What time is it? Do you have a light?) … It's also to be noted that such standardized questions… by virtue of the fact that they are standardized, provide for the relevance of the sequencing rules such that one knows when one of those questions is complete. Further, one knows what an answer to such a question looks like, so that the one who asked the question can know when the thing that stands as an answer will have finished, and thus provide that the other can talk again. For persons who don't know the discourse patterns of somebody they're dealing with, the use of standardized objects to build the beginning of a conversation may be quite important. [107] The person who is asking the questions seems to have first rights to perform an operation on the set of answers. You can call it 'draw a conclusion'. Socrates used the phrase 'add them up'. It was very basic this way of doing dialectic. He would go along and then say at some point, "Well, let's see where we are. Let's add up the answers and draw some conclusion." And it's that right that provides for a lot of what look like strugglings in some conversations, where the attempt to move into the position of 'questioner' seems to be quite a thing that persons try to do... As long as one is in the position of doing the questions, then in part they have control of the conversation.

reinforcement (see glances are actions; uncertainty)

rules [37] are not causally related but highly regulated phenomena [142] For example, asking for help is regulated. There are strong feelings - and strong maxims which provide those feelings - that you don't turn to "a stranger" for help with personal troubles. And you certainly don't as a first act. There's a proper sequence.

scapegoating (An individual is chosen by a group to be the sole recipient of blame for some wrongdoing and in doing this, the collective guilt of the group for that wrongdoing is purged) [93] If a member of a group does something which is sanctionable, then either that member or some other member of the group may be selected by some procedure … and a sanction is applied to that person or persons. That being done, guilt is purged

secrets and lies [146] a major device for children, for separating out [*Class I and Class II rules] are things like lies and secrets. They begin to discover that there are some things which they can violate, that, if the adult doesn't know, isn't told, doesn't find out about, nothing happens. And that may operate in alternative ways. One way it may operate is that they develop an adequate notion of reality; that is, they separate Class I and Class II rules. (See also neurotic)

sequencing [169] For the linguists, almost exclusively the largest unit of investigation, the largest unit they seek to describe, is a sentence… If we want to study natural activities in their natural sequences, we have to deal with, for example, the obvious fact that a sentence is not necessarily a 'complete utterance'… We want to construct some unit which will permit us to study actual activities … in the first place make of it 'a unit' - a natural unit and an analytic unit at the same time? The question then becomes, what do we need, to do that? First we need some rules of sequencing, and then some objects that will be handled by the rules of sequencing. Now, if we restrict our attention at the beginning, to two-party conversations, then we can get something extremely simple… (See also adequate complete utterance; conversation analysis)

society [159] A collection of many people who are not members of this or that *class, who are able to see what things are (They see by drawing on their tacit of explicit knowledge of the norms of their society. They can do this because, although society members are not members of all of society’s classes, they are nevertheless members of the ‘social metaclass’ that comprises of every individual who has been socialized into the perspective of a given society. This knowledge allows individuals to make inferences about other members society’s other classes: for example, individual men can make inferences about individual women who are members of the same society, even though ‘women’ are a different class). (See also category)

subversion (Adam's Problem) (A person’s inner character is observable in outer manifestations. One way that persons go about seeing a person’s inner moral character is by observing activities and explaining them by reference to some procedures which they take it properly occur as the activities occur.) [206] "And the Lord God called unto Adam, and said unto him, Where art thou? And he said… I was afraid, because I was naked; and I hid myself. [207] And He said, who told thee that thou was naked? Hast thou eaten of the tree whereof I commanded thee that thou shouldst not eat? And the man said… I did eat." (To illustrate what he means by subversion, Sacks’s gives the example of a boy Raymond who learns that his parents can tell that he washed his teeth by virtue of the appearance on his face of tooth powder [on his shirt]).

tautological statements (as anti MIR devices) [94] There is a class of [talk which on the surface] looks like tautological statements, 'Women will be women', 'White folks is white folks'… that is, that they say nothing. But given the use of *M.I.R. modifiers, these tautological statements seem to be - to use an awkward phrase - 'anti-modifier modifiers'. (they provide for the re-relevance of whatever it is that's known about the category. ) For example, "He's a Negro, but the things you can say about Negroes do apply to him."

uncertainty [163] There is a kind of a pattern [that informs the judgment of legitimate and illegitimate events]. The pattern is that there is (1) A normal way of going about things. (2) a subversive way of going about things with a normal appearance (3) and an uncertainty feature operating with regard to all events. [164] the question of uncertainty is in part handleable in the observations you've made in these reports; that is, that people check out the things they've noticed with somebody else … [161] since it's known that persons can be acting in a subversive fashion and appear to be acting normally, at the same time it's not known whether in fact they happen to be operating in a subversive fashion. And as a matter of fact it would be more of a problem to account for why that uncertainty factor is not operating, than to account for why it is operating when it is. [164] We have to consider is, what is it that they get out of checking out their noticings with somebody else? Some of you talk about "reinforcement"; one reinforces their determination that someone was doing something wrong, or that a combination was wrong, by making a check-out with somebody else. And thereby the norms are reinforced. Which is a nice, Durkheimian kind of argument, and it may be true. But there's something that has to be seen about that claim, and that is the mechanism of the procedure in the first place, which is what I'm interested in. (See also category; inference-making machine; order of depth; order of relevancy)