THE SOCIOLOGICAL IMAGINATION
C. WRIGHT MILLS

NEW YORK Oxford University Press 1959

Appendix

On Intellectual Craftsmanship
TO THE INDIVIDUAL social scientist who feels himself a part of the classic tradition, social science is the practice of a craft. A man at work on problems of substance, he is among those who are quickly made impatient and weary by elaborate discussions of method-and-theory-in-general; so much of it interrupts his proper studies. It is much better, he believes, to have one account by a working student of how he is going about his work than a dozen 'codifications of procedure' by specialists who as often as not have never done much work of consequence. Only by conversations in which experienced thinkers exchange information about their actual ways of working can a useful sense of method and theory be imparted to the beginning student. I feel it useful, therefore, to report in some detail how I go about my craft. This is necessarily a personal statement, but it is written with the hope that others, especially those beginning independent work, will make it less personal by the facts of their own experience.

1
It is best to begin, I think, by reminding you, the beginning student, that the most admirable thinkers within the scholarly community you have chosen to join do not split their work from their lives. They seem to take both too seriously to allow such dissociation, and they want to use each for the enrichment of the other. Of course, such a split is the prevailing convention among men in general, deriving, I suppose, from the hollowness of the work which men in general now do. But you will have recognized that as a scholar you have the exceptional opportunity of designing away of living which will encourage the habits of good workmanship. Scholarship is a choice of how to live as well as a choice of career; whether he knows it or not, the intellectual workman forms his own self as he works toward the perfection of his craft; to realize his own potentialities, and any opportunities that come his way, he constructs a character which has as its core the qualities of the good workman. What this means is that you must learn to use your life experience in your intellectual work: continually to examine and interpret it. In this sense craftsmanship is the center of yourself and you are personally involved in every intellectual product upon which you may work. To say that you can have experience, means, for one thing, that your past plays into and affects your present, and that it defines your capacity for future experience. As a social scientist, you have to control this rather elaborate interplay, to capture what you experience and sort it out; only in this way can you hope to use it to guide and test your reflection, and in the process shape yourself as an intellectual craftsman. But how can you do this? One answer is: you must set up a file, which is, I suppose, a sociologist's way of saying: keep a journal. Many creative writers keep journals; the sociologist's need for systematic reflection demands it. In such a file as I am going to describe, there is joined personal experience and professional activities, studies under way and studies planned. In this file, you, as an intellectual craftsman, will try to get together what you are doing intellectually and what you are experiencing as a person. Here you will not be afraid to use your experience and relate it directly to various work in progress. By serving as a check on repititious work, your file also enables you to conserve your energy. It also encourages you to capture `fringe-thoughts': various ideas which may be byproducts of everyday life, snatches of conversation overheard on the street, or, for that matter, dreams. Once noted, these may lead to more systematic thinking, as well as lend intellectual relevance to more directed experience. You will have often noticed how carefully accomplished thinkers treat their own minds, how closely they observe their development and organize their experience. The reason they treasure their smallest experiences is that, in the course of a lifetime, modern man has so very little personal experience and yet experience is so important as a source of original intellectual work. To be able to trust yet to be skeptical of your own experience, I have come to believe, is one mark of the mature workman. This ambiguous confidence is indispensable to originality in any intellectual pursuit, and the file is one way by which you can develop and justify such confidence. By keeping an adequate file and thus developing self-reflective habits, you learn how to keep your inner world awake. Whenever you feel strongly about events or ideas you must try not to let them pass from your mind, but instead to formulate them for your files and in so doing draw out their implications, show yourself either how foolish these feelings or ideas are, or how they might be articulated into productive shape. The file also helps you build up the habit of writing. You cannot `keep your hand in' if you do not write something at least every week. In developing the file, you can experiment as a writer and thus, as they say, develop your powers of expression. To maintain a file is to engage in the controlled experience. One of the very worst things that happens to social scientists is that they feel the need to write of their plans' on only one occasion: when they are going to ask for money for a specific piece of research or `a project' It is as a request for funds that most ‘planning’ is done, or at least carefully written about. However standard the practice, I think this very bad: It is bound in some degree to be salesmanship, and, given prevailing expectations, very likely to result in painstaking pretensions; the project is likely to be presented, rounded out in some arbitrary manner long before it ought to be; it is often a contrived thing, aimed at getting the money for ulterior purposes, however valuable, as well as for the research presented. A practicing social scientist ought periodically to review the state of my problems and plans. A young man, just at the beginning of his independent work, ought to

reflect on this, but he cannot be expected-and shouldn't expect himself-to get very far with it, and certainly he ought not to become rigidly committed to any one plan. About all he can do is line up his thesis, which unfortunately is often his first supposedly independent piece of work of any length. It is when you are about half-way through the time you have for work, or about one-third through, that such reviewing is most likely to be fruitful -and perhaps even of interest to others. Any working social scientist who is well on his way ought at all times to have so many plans, which is to say ideas, that the question is always, which of them am I, ought I, to work on next? And he should keep a special little file for his master agenda, which he writes and rewrites just for himself and perhaps for discussion with friends. From time to time he ought to review this very carefully and purposefully, and sometimes too, when he is relaxed. Some such procedure is one of the indispensable means by which your intellectual enterprise is kept oriented and under control. A widespread, informal interchange of such reviews of `the state of my problems' among working social scientists is, I suggest, the only basis for an adequate statement of the leading problems of social science. It is unlikely that in any free intellectual community there would be and certainly there ought not to be any ‘monolithic’ array of problems. In such a community, were it flourishing in a vigorous way, there would be interludes of discussion among individuals about future work. Three kinds of interludes-on problems, methods, theory-ought to come out of the work of social scientists, and lead into it again; they should be shaped by work-in-progress and to some extent guide that work. It is for such interludes that a professional association finds its intellectual reason for being. And for them too your own file is needed. Under various topics in your file there are ideas, personal notes, excerpts from books, bibliographical items and outlines of projects. It is, I suppose, a matter of arbitrary habit, but I think you will find it well to sort all these items into a master file of `projects,' with many subdivisions. The topics, of course, change, sometimes quite frequently. For instance, as a student working toward the preliminary examination, writing a thesis, and, at the same time, doing term papers, your files will be arranged in those three areas of endeavor: But after a year or so of graduate work, you will begin to re-organize the whole file in relation to the main project of your thesis. Then as you pursue your work you will notice that no one project ever dominates it, or sets the master categories in which it is arranged. In fact, the use of the file encourages expansion of the categories which you use in your thinking. And the way in which these categories change, some being dropped and others being added-is an index of your intellectual progress and breadth. Eventually, the files will come to be arranged according to several large projects, having many sub-projects that change from year to year. All this involves the taking of notes. You will have to acquire the habit of taking a large volume of notes from any worth-while book you read-although, I have to say, you may get better work out of yourself when you read really bad books. The first step in translating experience, either of other men's writing, or of your own life, into the intellectual sphere, is to give it form. Merely to name an item of experience often invites you to explain it; the mere taking of a note from a book is often a prod to reflection. At the same time, of course, the taking of a note is a great aid in comprehending what you are reading. Your notes may turn out, as mine do, to be of two sorts: in reading certain very important books you try to grasp the structure of the writer's argument, and take notes accordingly; but more frequently, and after a few years of independent work, rather than read entire books, you will very often read parts of many books from the point of view of some particular theme or topic in which you are interested and concerning which you have plans in your file. Therefore, you will take notes which do not fairly represent the books you read. You are using this particular idea, this particular fact, for the realization of your own projects.

2
But how is this file-which so far must seem to you more like a curious sort of `literary' journal-used in intellectual production? The maintenance of such a file is intellectual production. It is a continually growing store of facts and ideas, from the most vague to the most finished. For example, the first thing I did upon deciding on a study of the elite was to make a crude outline based on a listing of the types of people that I wished to understand. Just how and why I decided to do such a study may suggest one way in which one's life experiences feed one's intellectual work. I forget just when I became technically concerned with `stratification,' but I think it must have been on first reading Veblen. He had always seemed to me very loose, even vague, about his `business' and `industrial' employments, which are a kind of translation of Marx for the academic American public. At any rate, I wrote a book on labor organizations and labor leaders-a politically motivated task; then a book on the middle classes-a task primarily motivated by the desire to articulate my own experience in New York City since 1945. It was thereupon suggested by friends that I round out a trilogy by writing a book on the upper classes. I think the possibility had been in my mind; I had read Balzac off and on especially during the 'forties, and had been much taken with his self-appointed task of `covering all the major classes and types in the society of the era he wished to make his own. I had also written a paper on `The Business Elite,' and had collected and arranged statistics about the careers of the topmost men in American politics since the Constitution. These two tasks were primarily inspired by seminar work in American history. In doing these several articles and books and in preparing courses in stratification, there was of course a residue of ideas and facts about the upper classes. Especially in the study of social stratification is it difficult to avoid going beyond one's immediate subject, because the reality of any one stratum is in large part its relations to the rest. Accordingly, I began to think of a book on the elite. And yet that is not `really' how `the project' arose; what really happened is (1) that the idea and the plan came out of my files, for all projects with me begin and end with them, and books are simply organized releases from the continuous work that goes into them; ( 2 ) that after a while, the whole set of problems involved came to dominate me.

After making my crude outline I examined my entire file, not only those parts of it that obviously bore on my topic, but also those which seemed to have no relevance whatsoever. Imagination is often successfully invited by putting together hitherto isolated items, by finding unsuspected connections. I made new units in the file for this particular range of problems, which of course, led to new arrangements of other parts of the file. As you re-arrange a filing system, you often find that you are, as it were, loosening your imagination. Apparently this occurs by means of your attempt to combine various ideas and notes on different topics. It is a sort of logic of combination, and `chance' sometimes plays a curiously large part in it. In a relaxed way, you try to engage your intellectual resources, as exemplified in the file, with the new themes. In the present case, I also began to use my observations and daily experiences. I thought first of experiences I had had which bore upon elite problems, and then I went and talked with those who, I thought, might have experienced or considered the issues. As a matter of fact, I now began to alter the character of my routine so as to include in it (1) people who were among those whom I wanted to study, (2) people in close contact with them, and ( 3 ) people interested in them usually in some professional way. I do not know the full social conditions of the best intellectual workmanship, but certainly surrounding oneself by a circle of people who will listen and talk-and at times they have to be imaginary characters-is one of them. At any rate I try to surround myself with all the relevant environment-social and intellectual-that I think might lead me into thinking well along the lines of my work. That is one meaning of my remarks above about the fusion of personal and intellectual life. Good work in social science today is not, and usually cannot be, made up of one clear-cut empirical research. It is, rather, composed of a good many studies which at key points anchor general statements about the shape and the trend of the subject. So the decision-what are these anchor points?-cannot be made until existing materials are re-worked and general hypothetical statements constructed. Now, among `existing materials,' I found in the files three types relevant to my study of the elite: several theories having to do with the topic; materials already worked up by others as evidence for those theories; and materials already gathered and in various stages of accessible centralization, but not yet made theoretically relevant. Only after completing a first draft of a theory with the aid of such existing materials as these can I efficiently locate my own pivotal assertions and hunches and design researches to test them-and maybe I will not have to, although of course I know I will later have to shuttle back and forth between existing materials and my own research. Any final statement must not only `cover the data' so far as the data are available and known to me, but must also in some way, positively or negatively, take into account the available theories. Sometimes this `taking into account' of an idea is easily done by a simple confrontation of the idea with overturning or supporting fact; sometimes a detailed analysis or qualification is needed. Sometimes I can arrange the available theories systematically as a range of choices, and so allow their range to organize the problem itself." But sometimes I allow such theories to come up only in my own arrangement, in quite various contexts. At any rate, in the book on the elite I had to take into account the work of such men as Mosca, Schumpeter, Veblen, Marx, Lasswell, Michel, Weber, and Pareto. In looking over some of the notes on these writers, I find that they offer three types of statement: ( a ) from some, you learn directly by restating systematically what the man says on given points or as a whole; ( b ) some you accept or refute, giving reasons and arguments; ( c ) others you use as a source of suggestions for your own elaborations and projects. This involves grasping a point and then asking: How can I put this into testable shape, and how can I test it? How can I use this as a center from which to elaborate-as a perspective from which descriptive details emerge as relevant? It is in this handling of existing ideas, of course, that you feel yourself in continuity with previous work. Here are two excerpts from preliminary notes on Mosca, which may illustrate what I have been trying to describe: In addition to his historical anecdotes, Mosca backs up his thesis with this assertion: It's the power of organization that enables the minority always to rule. There are organized minorities and they run things and men. There are unorganized majorities and they are run. But: why not also consider (1) the organized minority, (2) the organized majority, (3) the unorganized minority, (4) the unorganized majority: This is worth full-scale exploration. The first thing that has to be straightened out: just what is the meaning of `organized'? I think Mosca means: capable of more or less continuous and co-ordinated policies and actions. If so, his thesis is right by definition. He would also say, I believe, that an ‘organized majority’ is impossible because all it would amount to is that new leaders, new -elites, would be on top of these majority organizations, and he is quite ready to pick up these leaders in his `The Ruling Class.' He calls them `directing minorities,' all of which is pretty flimsy stuff alongside his big statement. One thing that occurs to me (I think it is the core of the problems of definition that Mosca presents to us) is this: from the nineteenth to the twentieth century, we have witnessed a shift from a society organized as 1 and 4 to a society established more in terms of 3 and 2. We have moved from an elite state to an organization state, in which the elite is no longer so organized nor so unilaterally powerful, and the mass is more organized and more powerful. Some power has been made in the streets, and around it the whole social structures and their 'elites' have pivoted. And what section of the ruling class is more organized than the farm bloc? That's not a rhetorical question: I can answer it either way at this time; it's a matter of degree. All I want now is to get it out in the open. Mosca makes one point that seems to me excellent and worth elaborating further: There is often in `the ruling class,' according to him, a top clique and there is this second and larger stratum, with which (a) the top is in continuous and immediate contact, and with which (b) it shares ideas and sentiments and hence, he believes, policies. (page 430) Check and see if anywhere else in the

book, he makes other points of connection. Is the clique recruited largely from the second level? Is the top, in some way, responsible for, or at least sensitive to, this second stratum? Now forget Mosca: in another vocabulary, we have, (a) the elite, by which we here mean that top clique, (b) those who count, and (c) all the others. Membership in the second and third, in this scheme, is defined by the first, and the second may be quite varied in its size and composition and relations with the first and the third. (What, by the way, is the range of variations of the relations of (b) to (a) and to (c)? Examine Mosca for hints and further extend this by considering it systematically.) This scheme may enable me more neatly to take into account the different elites, which are elites according to the several dimensions of stratification. Also, of course, to pick up in a neat and meaningful way the Paretian distinction of governing and non-governing elites, in a way less formal than Pareto. Certainly many top-status people would at least be in the second. So would the big rich. The Clique or The Elite would refer to power, or to authority, as the case may be. The elite in this vocabulary would always mean the power elite. The other top people would be the upper classes or the upper circles. So in a way, maybe, we can use this in connection with two major problems: the structure of the elite; and the conceptual-later perhaps, the substantive-relations of stratification and elite theories. (Work this out.) From the standpoint of power, it is easier to pick out those who count than those who rule. When we try to do the first we select the top levels as a sort of loose aggregate and we are guided by position. But when we attempt the second, we must indicate in clear detail how they wield power and just how they are related to the social instrumentalities through which power is exercised. Also we deal more with persons than positions, or at least have to take persons into account. Now power in the United States involves more than one elite. How can we judge the relative positions of these several elites? Depends upon the issue and decisions being made. One elite sees another as among those who count. There is this mutual recognition among the elite, that other elites count; in one way or another they are important people to one another. Project: select 3 or 4 key decisions of last decade-to drop the atom, to cut or raise steel production, the G.M. strike of '45-and trace in detail the personnel involved in each of them. Might use `decisions' and decision-making as interview pegs when you go out for intensives.

3
There comes a time in the course of your work when you are through with other books. Whatever you want from them is down in your notes and abstracts; and on the margins of these notes, as well as in a separate file, are ideas for empirical studies. Now I do not like to do empirical work if I can possibly avoid it. If one has no staff it is a great deal of trouble; if one does employ a staff, then the staff is often even more trouble. In the intellectual condition of the social sciences today, there is so much to do by way of initial `structuring' ( let the word stand for the kind of work I am describing) that much `empirical research' is bound to be thin and uninteresting. Much of it, in fact, is a formal exercise for beginning students, and sometimes a useful pursuit for those who are not able to handle the more difficult substantive problems of social science. There is no more virtue in empirical inquiry as such than in reading as such. The purpose of empirical inquiry is to settle disagreements and doubts about facts, and thus to make arguments more fruitful by basing all sides more substantively. Facts discipline reason; but reason is the advance guard in any field of learning. Although you will never be able to get the money with which to do many of the empirical studies you design, it is necessary that you continue designing them. For once you lay out an empirical study, even if you do not follow it through, it leads you to a new search for data, which often turn out to have unsuspected relevance to your problems. Just as it is foolish to design a field study if the answer can be found in a library, it is foolish to think you have exhausted the books before you have translated them into appropriate empirical studies, which merely means into questions of fact. Empirical projects necessary to my kind of work must promise, first, to have relevance for the first draft, of which I wrote above; they have to confirm it in its original form or they have to cause its modification. Or to put it more pretentiously, they must have implications for theoretical constructions. Second, the projects must be efficient and neat and, if possible, ingenious. By this I mean that they must promise to yield a great deal of material in proportion to the time and effort they involve. But how is this to be done? The most economical way to state a problem is in such a way as to solve as much of it as possible by reasoning alone. By reasoning we try (a) to isolate each ques tion of fact that remains; ( b ) to ask these questions of fact in such ways that the answers promise to help us solve further problems by further reasoning.³To take hold of problems in this way, you have to pay attention to four stages; but it is usually best to go through all four many times rather than to get stuck in any one of them too long. The steps are: (1) the elements and definitions that, from your general awareness of the topic, issue, or area of concern, you think you are going to have to take into account; ( 2 ) the logical relations between these definitions and elements; building these little preliminary models, by the way, affords the best chance for the play of the sociological imagination; (3) the elimination of false views due to omissions of needed elements, improper or unclear definitions of terms, or undue emphasis on some part of the range and its logical extensions; (4) statement and re-statement of the questions of fact that remain. The third step, by the way, is a very necessary but often neglected part of any adequate statement of a problem. The popular awareness of the problem-the problem as an issue and as a trouble-must be carefully taken into account: that is part of the problem. Scholarly statements, of course, must be carefully ex amined and either used up in the re-statement being made, or thrown out.

Before deciding upon the empirical studies necessary for the job at hand, I began to sketch a larger design within which various small-scale studies began to aris e. Again, I excerpt from the files: I am not yet in a position to study the upper circles as a whole in a systematic and empirical way. So what I do is set forth some definitions and procedures that form a sort of ideal design for such a study. I can then attempt, first, to gather existing materials that approximate this design; second, to think of convenient ways of gathering materials, given the existing indices, that satisfy it at crucial points; and third, as I proceed, to make more specific the full-scale, empirical researches that would in the end be necessary. The upper circles must, of course, be defined systematically in terms of specific variables. Formally-and this is more or less Pareto's way-they are the people who have ‘the most’ of whatever is available of any given value or set of values. So I have to make two decisions: What variables shall I take as the criteria, and what do I mean by `the most'? After I've decided on my variables, I must construct the best indices I can, if possible quantifiable indices, in order to distribute the population in terms of them; only then can I begin to decide what I mean by ‘the most’. For this should, in part, be left for determination by empirical inspection of the various distributions, and their overlaps. My key variables should, at first, be general enough to give me some latitude in the choice of indices, yet specific enough to invite the search for empirical indices. As I go along, I'll have to shuttle between conceptions and indices, guided by the desire not to lose intended meanings and yet to be quite specific about them. Here are the four Weberian variables with which I will begin: 1. Class refers to sources and amounts of income. So I'll need property distributions and income distributions. The ideal material here (which is very scarce, and unfortunately dated) is a cross-tabulation of source and amount of annual income. Thus, we know that X per cent of the population received during 1938 Y millions or over, and that Z per cent of all this money was from property, W per cent from entrepreneurial withdrawal, Q per cent from wages and salaries. Along this class dimension, I can define the upper circles-those who have the most-either as those who receive given amounts of income during a given time-or, as those who make up the upper two per cent of the income pyramid. Look into treasury records and lists of big taxpayers. See if TNEC tables on source and amount of income can be brought up to date. II. Status refers to the amounts of deference received. For this, there are no simple or quantifiable indices. Existing indices require personal interviews for their application, are limited so far to local community studies, and are mostly no good anyway. There is the further problem that, unlike class, status involves social relations: at least one to receive and one to bestow the deference. It is easy to confuse publicity with deference-or rather, we do not yet know whether or not volume of publicity should be used as an index to status position, although it is the most easily available (For example: On one or two successive days in mid-March 1952, the following categories of people were mentioned by name in the New York Times-or on selected pages -work this out) III. Power refers to the realization of one's will even if others resist. Like status, this has not been well indexed. I don't think I can keep it a single dimension, but will have to talk (a) of formal authority defined by rights and powers of positions in various institutions, especially military, political, and economic. And (b) powers known informally to be exercised but not formally instituted-pressure group leaders, propagandists with extensive media at their disposal, and so on. IV. Occupation refers to activities that are paid for. Here, again, I must choose just which feature of occupation I should seize upon. (a) If I use the average incomes of various occupations, to rank them, I am of course using occupation as an index, and as the basis of, class. In like manner (b) if I use the status or the power typically attached to different occupations, then I am using occupations as indices, and bases, of power and skill or talent. But this is by no means an easy way to classify people. Skill-no more than status-is not a homogeneous something of which there is more or less. Attempts to treat it as such have usually been put in terms of the length of time required to acquire various skills, and maybe that will have to do, although I hope I can think of something better. Those are fine types of problems I will have to solve in order to define analytically and empirically the upper circles. in terms of these four key variables. For purposes of design, assume I have solved them to my satisfaction, and that I have distributed the population in terms of each of them. I would then have four sets of people; those at the top in class, status, power, and skill. Suppose further, that I had singled out the top two per cent of each distribution, as an upper circle. I then confront this empirically answerable question: How much, if any, overlap is there among each of these four distributions? One range of possibilities can be located within this simple chart: (+= top two per cent; - = lower 98 per cent).

CLASS +
STATUS

STATUS

+ Skill Power - Skill

+ + -

+ 1 5 9 13

2 6 10 14

+ 3 7 11 15

4 8 12 16

This diagram, if I had the materials to fill it, would contain major data and many important problems for a study of the upper circles. It would provide keys to many definitional and substantive questions. I don't have the data, and I shan't be able to get it-which makes it all the more important that I speculate about it, for in the course of such reflection, if it is guided by the desire to approximate the empirical requirements of an ideal design, I'll come upon important areas, on which I might be able to get materials that are relevant as anchor points and guides to further reflection. There are two additional points which I must add to this general model in order to make it formally complete. Full conceptions of upper strata require attention to duration and mobility. The task here is to determine positions (1-16) between which there is typical movement of individuals and groups-within the present generation, and among the last two or three generations. This introduces the temporal dimension of biography (or careerlines) and of history into the scheme. These are not merely further empirical questions; they are also definitionally relevant. For (a) we want to leave open whether or not in classifying people in terms of any of our key variables, we should define our categories in terms of how long they, or their families, have occupied the position in question. For example, I might want to decide that the upper two per cent of status-or at least one important type of status rank-consists of those up there for at least two generations. Also (b) I want to leave open the question of whether or not I should construct `a stratum' not only in terms of an intersection of several variables, but also, in line with Weber's neglected definition of `social class,' as composed of those positions between which there is `typical and easy mobility.' Thus, the lower white-collar occupations and middle and upper wageworker jobs in certain industries seem to be forming, in this sense, a stratum. In the course of the reading and analyzing of others' theories, designing ideal research, and perusing the files, you will begin to draw up a list of specific studies. Some of them are too big to handle, and will in time be regretfully given up; some will end as materials for a paragraph, a section, a sentence, a chapter; some will become pervading themes to be woven into an entire book. Here again are initial notes for several such projects: (1) A time-budget analysis of a typical working day of ten top executives of large corporations, and the same for ten federal administrators. These observations will be combined with detailed 'life history' interviews. The aim here is to describe the major routines and decisions, partly at least in terms o£ time devoted to them, and to gain an insight into the factors relevant to the decisions made. The procedure will naturally vary with the degree of co-operation secured, but ideally will involve first, an interview in which the life history and present situation of the man is made clear; second, observations of the day, actually sitting in a corner of the man's office, and following him around; third, a longish interview that evening or the next day in which we go over the whole day and probe the subjective processes involved in the external behavior we've observed. (2) An analysis of upper-class week ends, in which the routines are closely observed and followed by probing interviews with the man and other members of the family on the Monday following. For both these tasks I've fairly good contacts and of course good contacts, if handled properly, lead to better ones. [added 1957: this turned out to be an illusion.] (3) A study of the expense account and other privileges which, along with salaries and other incomes, form the standard and the style of living of the top levels. The idea here is to get something concrete on 'the bureaucratization of consumption,' the transfer of private ex penses to business accounts. (4) Bring up to date the type of information contained in such books as Lindbergh’s America's Sixty Families, which is dated as of the tax returns for 1923. (5) Gather and systematize, from treasury records and other government sources, the distribution of various types of private property by amounts held. (6) A career-line study of the Presidents, all cabinet members, and all members of the Supreme Court. This I already have on IBM cards from the Constitutional period through Tremens’ second term, but I want to expand the items used and analyze it afresh. There are other-some 35-projects of this sort (for example, comparison of the amounts of money spent in the presidential elections of 1896 and 1952, detailed comparison of Morgan of 1919 and Kaiser of 1950, and something concrete on the careers of ‘Admirals and Generals'). But, as one goes along, one must of course adjust his aim to what is accessible.

After these designs were written down, I began to read historical works on top groups, taking random ( and unfiled ) notes and interpreting the reading. You do not really have to study a topic you are working on; for as I have said, once you are into it, it is everywhere. You are sensible to its themes; you see and hear them everywhere in your experience, especially, it always seems to me, in apparently unrelated areas. Even the mass media, especially bad movies and cheap novels and picture magazines and night radio, are disclosed in fresh importance to you.

4
But, you may ask, how do ideas come? How is the imagination spurred to put all the images and facts together, to make images relevant and lend meaning to facts? I do not think I can really answer that; all I can do is talk about the general conditions and a few simple techniques which have seemed to increase my chances to come out with something. The sociological imagination, I remind you, in considerable part consists of the capacity to shift from one perspective to another, and in the process to build up an adequate view of a total society and of its components. It is this imagination, of course, that sets off the social scientist from the mere technician. Adequate technicians can be trained in a few years. The sociological imagination can also be cultivated; certainly it seldom occurs without a great deal of often routine work. Yet there is an unexpected quality about it, perhaps because its essence is the combination of ideas that no one expected were combinable say, a mess of ideas from German philosophy and British economics. There is a playfulness of mind back of such combining as well as a truly fierce drive to make sense of the world, which the technician as such usually lacks. Perhaps he is too well trained, too precisely trained. Since one can be trained only in what is already known, training sometimes incapacitates one from learning new ways; it makes one rebel against what is bound to be at first loose and even sloppy. But you must cling to such vague images and notions, if they are yours, and you must work them out. For it is in such forms that original ideas, if any, almost always first appear. There are definite ways, I believe, of stimulating the sociological imagination: (1) On the most concrete level, the re-arranging of the file, as I have already said, is one way to invite imagination. You simply dump out heretofore disconnected folders, mixing up their contents, and then re-sort them. You try to do it in a more or less relaxed way. How often and how extensively you re-arrange the files will of course vary with different problems and with how well they are developing. But the mechanics of it are as simple as that. Of course, you will have in mind the several problems on which you are actively working, but you will also try to be passively receptive to unforeseen and unplanned linkages. (2) An attitude of playfulness toward the phrases and words with which various issues are defined often loosens up the imagination. Look up synonyms for each of your key terms in dictionaries as well as in technical books, in order to know the full range of their connotations. This simple habit will prod you to elaborate the terms of the problem and hence to define them less wordily and more precisely. For only if you know the several meanings which might be given to terms or phrases can you select the exact ones with which you want to work. But such an interest in words goes further than that. In all work, but especially in ex amining theoretical statements, you will try to keep close watch on the level of generality of every key term, and you will often find it useful to break down a high-level statement into more concrete meanings. When that is done, the statement often falls into two or three components, each lying along different dimensions. You will also try to move up the level of generality: remove the specific qualifiers and examine the re-formed statement or inference more abstractly, to see if you can stretch it or elaborate it. So from above and from below, you will try to probe, in search of clarified meaning, into every aspect and implication of the idea. (3) Many of the general notions you come upon, as you think about them, will be cast into types. A new classification is the usual beginning of fruitful developments. The skill to make up types and then to search for the conditions and consequences of each type will, in short, become an automatic procedure with you. Rather than rest content with existing classifications, in particular, common-sense ones, you will search for their common denominators and for differentiating factors within and between them. Good types require that the criteria of classification be explicit and systematic. To make them so you must develop the habit of cross-classification. The technique of cross-classifying is not of course limited to quantitative materials; as a matter of fact, it is the best way to imagine and to get hold of new types as well as to criticize and clarify old ones. Charts, tables, and diagrams of a qualitative sort are not only ways to display work already done; they are very often genuine tools of production. They clarify the `dimensions' of the types, which they also help you to imagine and build. As a matter of fact, in the past fifteen years, I do not believe I have written more than a dozen pages first-draft without some little cross-classification-although, of course, I do not always or even usually display such diagrams. Most of them flop, in which case you have still learned something. When they-work, they help you to think more clearly and to write more explicitly. They enable you to discover the range and the full relationships of the very terms with which you are thinking and of the facts with which you are dealing. For a working sociologist, cross-classification is what diagramming a sentence is for a diligent grammarian. In many ways, cross-classification is the very grammar of the sociological imagination. Like all grammar, it must be controlled and not allowed to run away from its purposes. (4) Often you get the best insights by considering extremes by thinking of the opposite of that with which you are directly concerned. If you think about despair, then also think about elation; if you study the miser, then also the spendthrift. The hardest thing in the world is to study one object; when you try to contrast objects, you get a better grip on the materials and you can then sort out the dimensions in terms of which the comparisons are made. You will find that shuttling between attention to these dimensions and to the concrete types is very illuminating. This technique is also logically sound, for without a sample, you can only guess about statistical frequencies anyway: what you can do is to give the range and the major types of some phenomenon,

and for that it is more economical to begin by constructing `polar types,' opposites along various dimensions. This does not mean, of course, that you will not strive to gain and to maintain a sense of proportion-to look for some lead to the frequencies of given types. One continually tries, in fact, to combine this quest with the search for indices for which one might find or collect statistics. The idea is to use a variety of viewpoints: you will, for instance, ask yourself how would a political scientist whom you have recently read approach this, and how would that experimental psychologist, or this historian? You try to think in terms of a variety of viewpoints and in this way to let your mind become a moving prism catching light from as many angles as possible. In this connection, the writing of dialogues is often very useful. You will quite often find yourself thinking against something, and in trying to understand a new intellectual field, one of the first things you might well do is to lay out the major arguments. One of the things meant by `being soaked in the literature' is being able to locate the opponents and the friends of every available viewpoint. By the way, it is not well to be too `soaked in the literature'; you may drown in it, like Mortimer Adler. Perhaps the point is to know when you ought to read, and when you ought not to. (5) The fact that, for the sake of simplicity, in cross-classification, you first work in terms of yes-or-no, encourages you to think of extreme opposites. That is generally good, for qualitative analysis cannot of course provide you with frequencies or magnitudes. Its technique and its end is to give you the range of types. For many purposes you need no more than that, although for some, of course, you do need to get a more precise idea of the proportions involved. The release of imagination can sometimes be achieved by deliberately inverting your sense of proportion.5 If something seems very minute, imagine it to be simply enormous, and ask yourself: What difference might that make? And vice versa, for gigantic phenomena. What would pre-literate villages look like with populations of 30 millions? Nowadays at least, I should never think of actually counting or measuring anything, before I had played with each of its elements and conditions and consequences in an imagined world in which I control the scale of everything. This is one thing statisticians ought to mean, but never seem to, by that horrible little phrase about ‘knowing the universe before you sample it’. ( 6) Whatever the problem with which you are concerned, you will find it helpful to try to get a comparative grip on the materials. The search for comparable cases, either in one civilization and historical period or in several, gives you leads. You would never think of describing an institution in twentieth-century America without trying to bear in mind similar institutions in other types of structures and periods. That is so even if you do not make explicit comparisons. In time you will come almost automatically to orient your reflection historically. One reason for doing so is that often what you are examining is limited in number: to get a comparative grip on it, you have to place it inside an historical frame. To put it another way, the contrasting type approach often requires the examination of historical materials. This sometimes results in points useful for a trend analysis, or it leads to a typology of phases. You will use historical materials, then, because of the desire for a fuller range, or for a more convenient range of some phenomenon-by which I mean a range that includes the variations along some known set of dimensions. Some knowledge of world history is indispensable to the sociologist; without such knowledge, no matter what else he knows, he is simply crippled. (7) There is, finally, a point which has more to do with the craft of putting a book together than with the release of the imagination. Yet these two are often one: how you go about arranging materials for presentation always affects the content of your work. The idea I have in mind I learned from a great editor, Lambert Davis, who, I suppose, after seeing what I have done with it, would not want to acknowledge it as his child. It is the distinction between theme and topic. A topic is a subject, like ‘the careers of corporation executives’ or ‘the increased power of military officials’ or ‘the decline of society matrons’. Usually most of what you have to say about a topic can readily be put into one chapter or a section of a chapter. But the order in which all your topics are arranged often brings you into the realm of themes. A theme is an idea, usually of some signal trend, some master conception, or a key distinction, like rationality and reason, for example. In working out the construction of a book, when you come to realize the two or three, or, as the case may be, the six or seven themes, then you will know that you are on top of the job. You will recognize these themes because they keep insisting upon being dragged into all sorts of topics and perhaps you will feel that they are mere repetitions. And sometimes that is all they are! Certainly very often they will be found in the more clotted and confused, the more badly written, sections of your manuscript. What you must do is sort them out and state them in a general way as clearly and briefly as you can. Then, quite systematically, you must cross-classify them with the full range of your topics. This means that you will ask of each topic: Just how is it affected by each of these themes? And again: Just what is the meaning, if any, for each of these themes of each of the topics? Sometimes a theme requires a chapter or a section for itself, perhaps when it is first introduced or perhaps in a summary statement toward the end. In general, I think most writers-as well as most systematic thinkers-would agree that at some point all the themes ought to appear together, in relation to one another. Often, although not always, it is possible to do this at the beginning of a book. Usually, in any well-constructed book, it must be done near the end. And, of course, all the way through you ought at least to try to relate the themes to each topic. It is easier to write about this than to do it, for it is usually not so mechanical a matter as it might appear. But sometimes it is -at least if the themes are properly sorted out and clarified. But that, of course, is the rub. For what I have here, in the context of literary craftsmanship, called themes, in the context of intellectual work are called ideas.

Sometimes, by the way, you may find that a book does not really have any themes. It is just a string of topics, surrounded, of course, by methodological introductions to methodology, and theoretical introductions to theory. These are indeed quite indispensable to the writing of books by men without ideas. And so is lack of intelligibility.

5
I know you will agree that you should present your work in as clear and simple language as your subject and your thought about it permit. But as you may have noticed, a turgid and polysyllabic prose does seem to prevail in the social sciences. I suppose those who use it believe they are imitating `physical science,' and are not aware that much of that prose is not altogether necessary. It has in fact been said with authority that there is `a serious crisis in literacy'-a crisis in which social scientists are very much involved.6 Is this peculiar language due to the fact that profound and subtle issues, concepts, methods, are being discussed? If not, then what are the reasons for what Malcolm Cowley aptly calls `socspeak'? 7 Is it really necessary to your proper work? If it is, there is nothing you can do about it; if it is not, then how can you avoid it? Such lack of ready intelligibility, I believe, usually has little or nothing to do with the complexity of subject matter, and nothing at all with profundity of thought. It has to do almost entirely with certain confusions of the academic writer about his own status. In many academic circles today anyone who tries to write in a widely intelligible way is liable to be condemned as a `mere literary man.' or, worse still, `a mere journalist' Perhaps you have already learned that these phrases, as commonly used, only indicate the spurious inference: superficial because readable. The academic man in America is trying to carry on a serious intellectual life in a social context that often seems quite set against it. His prestige must make up for many of the dominant values he has sacrificed by choosing an academic career. His claims for prestige readily become tied to his self-image as a ‘scientist’. To be called a `mere journalist' makes him feel undignified and shallow. It is this situation, I think, that is often at the bottom of the elaborate vocabulary and involved manner of speaking and writing. It is less difficult to learn this manner than not. It has become a convention-those who do not use it are subject to moral disapproval. It may be that it is the result of an academic closing of the ranks on the part of the mediocre, who understandably wish to exclude those who win the attention of intelligent people, academic and otherwise. To write is to raise a claim for the attention of readers. That is part of any style. To write is also to claim for oneself at least status enough to be read. The young academic man is very much involved in both claims, and because he feels his lack of public position, he often puts the claim for his own status before his claim for the attention of the reader to what he is saying. In fact, in America, even the most accomplished men of knowledge do not have much status among wide circles and publics. In this respect, the case of sociology has been an extreme one: in large part sociological habits of style stem from the time when sociologists had little status even with other academic men. Desire for status is one reason why academic men slip so readily into unintelligibility. And that, in turn, is one reason why they do not have the status they desire. A truly vicious circle-but one out of which any scholar can easily break. To overcome the academic prose you have first to overcome the academic pose. It is much less important to study grammar and Anglo-Saxon roots than to clarify your own answers to these three questions: (1) How difficult and complex after all is my subject? (2) When I write, what status am I claiming for myself? ( 3 ) For whom am I trying to write? (1) The usual answer to the first question is: Not so difficult and complex as the way in which you are writing about it. Proof of that is everywhere available: it is revealed by the ease with which 95 per cent of the books of social science can be translated into English.8 But, you may ask, do we not sometimes need technical terms? 9 Of course we do, but `technical' does not necessarily mean difficult, and certainly it does not mean jargon. If such technical terms are really necessary and also clear and precise, it is not difficult to use them in a context of plain English and thus introduce them meaningfully to the reader. Perhaps you may object that the ordinary words of common usage are often’ loaded’ with feelings and values, and that accordingly it might be well to avoid them in favor of new words or technical terms, here is my answer: it is true that ordinary words are often so loaded. But many technical terms in common use in social science are also loaded. To write clearly is to control these loads, to say exactly what you mean in such a way that this meaning and only this will be understood by others. Assume that your intended meaning is circumscribed by a six-foot circle, in which you are standing; assume that the meaning understood by your reader is another such circle, in which he is standing. The circles, let us hope, do overlap. The extent of that overlap is the extent of your communication. In the reader's circle the part that does not overlap-that is one area of uncontrolled meaning: he has made it up. In your circle the part that does not overlap-that is another token of your failure: you have not got it across. The skill of writing is to get the reader's circle of meaning to coincide exactly with yours, to write in such a way that both of you stand in the same circle of controlled meaning. My first point, then, is that most `socspeak' is unrelated to any complexity of subject matter or thought. It is used-I think almost entirely-to establish academic claims for one's self; to write in this way is to say to the reader ( often I am sure without knowing it): I know something that is so difficult you can understand it only if you first learn my difficult language. In the meantime, you are merely a journalist, a layman, or some other sort of underdeveloped type.'

(2) To answer the second question, we must distinguish two ways of presenting the work of social science according to the idea the writer has of himself, and the voice with which he speaks. One way results from the idea that he is a man who may shout, whisper, or chuckle-but who is always there. It is also clear what sort of man he is: whether confident or neurotic, direct or involuted, he is a center of experience and reasoning; now he has found out something, and he is telling us about it, and how he found it out. This is the voice behind the best expositions available in the English language. The other way of presenting work does not use any voice of any man. Such writing is not a `voice' at all. It is an autonomous sound. It is a prose manufactured by a machine. That it is full of jargon is not as noteworthy as that it is strongly mannered: it is not only impersonal; it is pretentiously impersonal. Government bulletins are sometimes written in this way. Business letters also. And a great deal of social science. Any writing-perhaps apart from that of certain truly great stylists-that is not imaginable as human speech is bad writing. ( 3 ) But finally there is the question of those who are to hear the voice-thinking about that also leads to characteristics of style. It is very important for any writer to have in mind just what kinds of people he is trying to speak to-and also what he really thinks of them. These are not easy questions: to answer them well requires decisions about oneself as well as knowledge of reading publics. To write is to raise a claim to be read, but by whom? One answer has been suggested by my colleague, Lionel Truing, who has given me permission to pass it on. You are to assume that you have been asked to give a lecture on some subject you know well, before an audience of teachers and students from all departments of a leading university, as well as an assortment of interested people from a near-by city. Assume that such an audience is before you and that they have a right to know; assume that you want to let them know. Now write. There are some four broad possibilities available to the social scientist as a writer. If he recognizes himself as a voice and assumes that he is speaking to some such public as I have indicated, he will try to write readable prose. If he assumes he is a voice but is not altogether aware of any public, he may easily fall into unintelligible ravings. Such a man had better be careful. If he considers himself less a voice than an agent of some impersonal sound, then-should he find a public-it will most likely be a cult. If, without knowing his own voice, he should not find any public, but speaks solely for some record kept by no one, they I suppose we have to admit that he is a true manufacturer of the standardized prose: an autonomous sound in a great empty hall. It is all rather frightening, as in a Kafka novel, and it ought to be: we have been talking about the edge of reason. The line between profundity and verbiage is often delicate, even perilous. No one should deny the curious charm of those who-as in Whitman's little poem-beginning their studies, are so pleased and awed by the first step that they hardly wish to go farther. Of itself, language does form a wonderful world, but, entangled in that world, we must not mistake the confusion of beginnings with the profundity of finished results. As a member of the academic community you should think of yourself as a representative of a truly great language, and you should expect and demand of yourself that when you speak or write you try to carry on the discourse of civilized man. There is one last point, which has to do with the interplay of writing and thinking. If you write solely with reference to what Hans Reichenbach has called the `context of discovery' you will be understood by very few people; moreover you will tend to be quite subjective in statement. To make whatever you think more objective, you must work in the context of presentation. At first, you `present' your thought to yourself, which is often called `thinking clearly.' Then when you feel that you have it straight, you present it to others-and often find that you have not made it clear. Now you are in the `context of presentation: Sometimes you will notice that as you try to present your thinking, you will modify it-not only in its form of statement but often in its content as well. You will get new ideas as you work in the context of presentation. In short, it will become a new context of discovery, different from the original one, on a higher level I think, because more socially objective. Here again, you cannot divorce how you think from how you write. You have to move back and forth between these two contexts, and whenever you move it is well to know where you might be going.

6
From what I have said, you will understand that in practice you never `start working on a project'; you are already `working,' either in a personal vein, in the files, in taking notes after browsing, or in guided endeavors. Following this way of living and working, you will always have many topics that you want to work out further. After you decide on some 'release,' you will try to use your entire file, your browsing in libraries, your conversation, your selections of people-all for this topic or theme. You are trying to build a little world containing all the key elements which enter into the work at hand, to put each in its place in a systematic way, continually to readjust this framework around developments in each part of it. Merely to live in such a constructed world is to know what is needed: ideas, facts, ideas, figures, ideas. So you will discover and describe, setting up types for the ordering of what you have found out, focusing and organizing experience by distinguishing items by name. This search for order will cause you to seek patterns and trends, to find relations that may be typical and causal. You will search, in short, for the meanings of what you come upon, for what may be interpreted as a visible token of something else that is not visible. You will make an inventory of everything that seems involved in whatever you are trying to understand; you will pare it down to essentials; then carefully and systematically you will relate these items to one

another in order to form a sort of working model. And then you will relate this model to whatever it is you are trying to explain. Sometimes it is that easy; often it just will not come. But always, among all the details, you will be searching for indicators that might point to the main drift, to the underlying forms and tendencies of the range of society in the middle of the twentieth century. For, in the end, it is this -the human variety-that you are always writing about. Thinking is a struggle for order and at the same time for com prehensiveness. You must not stop thinking too soon-or you will fail to know all that you should; you cannot leave it to go on forever, or you yourself will burst. It is this dilemma, I suppose, that makes reflection, on those rare occasions when it is more or less successful, the most passionate endeavor of which the human being is capable. Perhaps I can best summarize what I have been trying to say in the form of a few precepts and cautions: ( 1 ) Be a good craftsman: Avoid any rigid set of procedures. Above all, seek to develop and to use the sociological imagination. Avoid the fetishism of method and technique. Urge the rehabilitation of the unpretentious intellectual craftsman, and try to become such a craftsman yourself. Let every man be his own methodologist; let every man be his own theorist; let theory and method again become part of the practice of a craft. Stand for the primacy of the individual scholar; stand opposed to the ascendancy of research teams of technicians. Be one mind that is on its own confronting the problems of man and society. ( 2 ) Avoid the Byzantine oddity of associated and disassociated Concepts, the mannerism of verbiage. Urge upon yourself and upon others the simplicity of clear statement. Use more elaborated terms only when you believe firmly that their use enlarges the scope of your sensibilities, the precision of your references, the depth of your reasoning. Avoid using unintelligibility as a means of evading the making of judgments upon society-and as a. means of escaping your readers' judgments upon your own work. ( 3 ) Make any trans-historical constructions you think your work requires; also delve into sub-historical minutiae. Make up quite formal theory and build models as well as you can. Examine in detail little facts and their relations, and big unique events as well. But do not be fanatic: relate all such work, continuously and closely, to the level of historical reality. Do not assume that somebody else will do this for you, sometime, somewhere. Take as your task the defining of this reality; formulate your problems in its terms; on its level try to solve these problems and thus resolve the issues and the troubles they incorporate. And never write more than three pages without at least having in mind a solid example. ( 4 ) Do not study merely one small milieu after another; study the social structures in which milieu are organized. In terms of these studies of larger structures, select the milieux you need to study in detail, and study them in such a way as to understand the interplay of milieux with structure. Proceed in a similar way in so far as the span of time is concerned. Do not be merely a journalist, however precise a one. Know that journalism can be a great intellectual endeavor, but know also that yours is greater! So do not merely report minute researches into static knife-edge moments, or very short-term runs of time. Take as your time-span the course of human history, and locate within it the weeks, years, epochs you examine. ( 5 ) Realize that your aim is a fully comparative understanding of the social structures that have appeared and that do now exist in world history. Realize that to carry it out you must avoid the arbitrary specialization of prevailing academic departments. Specialize your work variously, according to topic, and above all according to significant problem. In formulating and in trying to solve these problems, do not hesitate, indeed seek, continually and imaginatively, to draw upon the perspectives and materials, the ideas and methods, of any and all sensible studies of man and society. They are your studies; they are part of what you are a part of; do not let them be taken from you by those who would close them off by weird jargon and pretensions of expertise. ( 6 ) Always keep your eyes open to the image of man-the generic notion of his human nature-which by your work you are assuming and implying; and also to the image of history-your notion of how history is being made. In a word, continually work out and revise your views of the problems of history, the problems of biography, and the problems of social structure in which biography and history intersect. Keep your eyes open to the varieties of individuality, and to the modes of epochal change. Use what you see and what you imagine, as the clues to your study of the human variety. (7) Know that you inherit and are carrying on the tradition of classic social analysis; so try to understand man not as an isolated fragment, not as an intelligible field or system in and of itself. Try to understand men and women as historical and social actors, and the ways in which the variety of men and women are intricately selected and intricately formed by the variety of human societies. Before you are through with any piece of work, no matter how indirectly on occasion, orient it to the central and continuing task of understanding the structure and the drift, the shaping and the meanings, of your own period, the terrible and magnificent world of human society in the second half of the twentieth century. ( 8 ) Do not allow public issues as they are officially formulated, or troubles as they are privately felt, to determine the problems that you take up for study. Above all, do not give up your moral and political autonomy by accepting in somebody else's terms the illiberal practicality of the bureaucratic ethos or the liberal practicality of the moral scatter. Know that many personal troubles cannot be solved merely as troubles, but must be understood in terms of public issues -and in terms of the problems of historymaking. Know that the human meaning of public issues must be revealed by relating them to personal troubles-and to the problems of the individual life. Know that the problems of social science, when adequately formulated, must include both troubles

and issues, both biography and history, and the range of their intricate relations. Within that range the life of the individual and the making of societies occur; and within that range the sociological imagination has its chance to make a difference in the quality of human life in our time.

FOOTNOTES
1

See, for example, Mills, White Collar, New York, Oxford University Press, 1951, chapter 13. I did the same kind of thing, in my notes, with Lederer and Gasset vs `elite theorists' as two reactions to eighteenth- and nineteenth century democratic doctrine.
2

There are also statements in Mosca about psychological laws supposed to support his view. Watch his use of the word `natural.' But this isn't central, and in addition, it's not worth considering. ³ Perhaps I ought to say the same things in a more pretentious language, in order to make evident to those who do not know, how important all this may be, to wit: Problematic situations have to be formulated with due attention to their theoretical and conceptual implications, and also to appropriate paradigms of empirical research and suitable models of verification. These paradigms and models in turn, must be so constructed that they permit further theoretical and conceptual implications to be drawn from their employment. The theoretical and conceptual implications of problematic situations should first be fully explored. To do this requires the social scientist to specify each such implication and consider it in relation to every other one, but also in such a way that it fits the paradigms of empirical research and the models of verification.
4

See the excellent articles on ‘insight’ and ‘creative endeavor’ by Hutchinson in Study o f Interpersonal Relations, edited by Patrick Mullahy, New York, Nelson, 1949.
5

By the way, some of this is what Kenneth Burke, in discussing Nietzsche, has called ‘perspective by incongruity’. See, by all means, Burke, Permanence and Change, New York, New Republic Books, 1938.
6

By Edmund Wilson, widely regarded as ’the best critic in the English speaking world,' who writes: ‘As for my experience with articles by experts in anthropology and sociology, it has led me to conclude that the requirement, in my ideal university, of having the papers in every department passed by a professor of English might result in revolutionizing these subjects-if indeed the second of them survived at all.’ A Piece of My Mind, New York, Farrar, Straus and Cudahy, 1956, p. 184.
7 8

Malcolm Cowley, `Sociological Habit Patterns in Linguistic Transmogrification; The Reporter, 20 September 1958, pp. 41 ff.

For some examples of such translation see above: chapter 2. By the way, on writing, the best book I know is Robert Graves and Alan Hodge, The Reader Over Your Shoulder, New York, Macmillan, 1944. See also the excellent discussions by Barzun and Graff, The Modern Researcher, op, cit., G. E. Montague, A Writer's Notes on His Trade, London, Pelican Books, 1930-1949, and Bonamy Dobrée, Modern Prose Style, Oxford, The Clarendon Press, 1934-50.
9

Those who understand mathematical language far better than I tell me that it is precise, economical, clear. That is why I am so suspicious of many social scientists who claim a central place for mathematics among the methods of social study but who write prose imprecisely, uneconomically, and unclearly. They should take a lesson from Paul Lazarsfeld, who believes in mathematics, very much indeed, and whose prose always reveals, even in first draft, the mathematical qualities indicated. When I cannot understand his mathematics, I know that it is because I am too ignorant; when I disagree with what he writes in non-mathematical language, I know it is because he is mistaken, for one always knows just what he is saying and hence just where he has gone wrong.