Theoretical Writing1

Barney G. Glaser, Ph.D., Hon. Ph.D.

Theoretical sorting has brought the analyst to the point of
pent-up pressure to write: to see the months of work actualized in
a “piece.” But this is only a personal pressure. The goal of
grounded theory methodology, above all is to offer the results to
the public, usually through one or more publications. We will
focus on writing for publication, which is the most frequent way
that the analyst can tell how people are “buying” what really
matters in sociology, or in other fields.

Both feedback on and use of publications will be the best
evaluation of the analyst’s grounded theory. It will be his main
source or criticism, constructive critique, and frequently of career
rewards. In any case, he has to write to expand his audience
beyond the limited number of close colleagues and students.
Unless there is a publication, his work will be relegated to limited
discussion, classroom presentation, or even private fantasy. The
rigor and value of grounded theory work deserves publication.
And many analysts have a stake in effecting wider publics, which
makes their substantive grounded theory count.

The best form to publish in sociology is through a
monograph. The highest rewards, in general, go for writing books,
for they probably reach the most diverse publics with the
maximum amount of material. Journal articles, of course, run a
close second. One solution which many analysts take is to write
chapters into articles, while fewer combine chapters into books.
We shall mainly focus here on chapter form, which is similar to
the article form with minor adjustments.

In this is the final stage of grounded theory methodology,
writing is a “write up” of piles of ideas from theoretical sorting.
Writing techniques are, perhaps, not as crucial as the techniques
characteristic of the previous stages, but they still crucial.

Since writing sums up the preceding work, it cannot be left
uncontrolled, perhaps to scuttle it. Rather, writing must capture
it. It must put into relief the conceptual work and its integration
into a theoretical explanation. So very often in qualitative
research, the theory is left implicit in the write-up as the analyst
gets caught up in the richness of the data.

Below we shall discuss the logic of construction, of shape and
of conceptual style of a monograph and a chapter. Then we
discuss the reworking of initial drafts, in order to sharpen the
shape and style. We briefly indicate our view of uses of the
literature, and close with recommendations for the analyst’s
theoretical pacing.

It must be underlined that the write-up of sorts is a theory of
a core variable which freezes the on-going for the moment. It is
unfortunate, perhaps, that writing has this “slice of reality”
character. We have covered this problem as best as possible by
using concepts and processes that have duration and are
independent of time and place. We also construct a theory that is
readily modifiable. The analyst should underscore these points in
his writing, because his writing probably will be read mainly as a
fixed conceptual description, not explanation, by most readers.
We are in essence stuck with this paradox.

Logic of Construction

Typically sociological monographs are constructed on the
basis of a “little logic.” It is the main building idea of the book,
hence the ensuing chapters. The little logic usually consists of no
more than a paragraph or two, and often just one long sentence.
In monographs it may be stated as an interest, a general idea, a
logical derivation, a hypothesis, a finding to be explored, an
explanation, a statement of purpose and so forth. In our case the
little logic states that the core variable explains a large amount of
the variation in a behavior or set of behaviors. For example, in
Awareness of Dying, we stated that awareness contexts account
for much of the behavior around a dying patient in the hospital.

These little logics are found in the preface, introduction,
editor’s note (when the author does not state it) or appendix.
Separate little logics may introduce each chapter, based on the
build up of the book. Or they may end a chapter to set up the
reason for the coming chapter. Sometimes each chapter further
refines the logic.

Implied in the little logic of monographs are many aspects
and assumptions of its construction. It implies whether the study
will be descriptive, verificational or focus on theory generation.
The little logic for a grounded theory monograph must clearly
reflect its generative intent. It also should imply the book’s
methodology, the book’s unity as a whole and its level of
conceptualization. It brings out the model for its integration; such
as in a grounded theory book, we state that the core variable will
explain a behavior, implying that it will be written this way as its
purpose. The little logic also brings out the unsolved question or
problem with its necessary dissonance, which will interest the
reader in finding out how the BSP [basic social process] will
process or resolve it. The little logic can be substantively coded or
theoretically coded, but is usually the former with the latter
implied.

In most monographs, we usually find one little logic and
sometimes two or none at all. A single one is all that is needed in
grounded theory, for it is based on a core variable analysis. Books
without any wander all over and books with two, as noted earlier,
find difficulty in handling both together adequately. The promise
implied in the little logic is one criteria by which to judge its
success: “Did her pull it off?” as the saying goes. The grounded
theorist should be cautious in his promise to the reader. The
modesty of his effort, should be underscored, but with no apology.

Some sociologists are noted for using the same little logic for
all their monographs and articles. It is usually a general
substantive coded one, such as the relationship of professionals to
client behavior, or the study of occupational careers, or the
organizational effects on political life, and so forth. These authors
have one theoretical code which they pursue as their “little logic”
in many different studies. In our case the little logic of the study
genuinely emerges.

Sometimes an author will overgeneralize his logic and spend
much of his time book specifying it. Others will state their logics
too specifically and soon transcend and leave them behind. The
reader then feels lost in trying to find the book he had been
invited to read. In our case, the level of generality of the logic is
based on the core category; hence the logic is consistent with the
level or conceptualization of the ensuing analysis.

Implied in the above discussion is a basic assumption of
grounded theory. Writing is a careful, systematic “construction
job”. It does not merely flow from a witty mind, no matter how
much wit might help. Readers who wish to write grounded theory
should look at several monographs to discover their little logics
and their properties. Such experience gives an armamentarium of
ideas on how to write a monograph effectively without
committing the errors of colleagues. This study is invaluable. It is
not to evaluate the substantive or abstract worth of monographs;
it is to learn more techniques in the construction of a book. For
example, one discovery we and our students have made is that
there are a number of authors who write a little logic with
minimal awareness of its import. Hence they are not or only
slightly constrained in following its implications for the ensuing
work. In grounded theory a little logic is written realistically and
with awareness so that it can be followed throughout the book.

Shape

In grounded theory we follow the standard shaping of
sociologically monographs and chapters or articles. For chapters
we being with an introduction which includes first, the general
problem, second, the methodology (if appropriate as in an article
or introduction to a book) and third, a prose outline of the coming
substantive theory for the chapter sections. Then we give the
sections. If the chapter is an introduction to a book we close with
the outline of the book. If it is a subsequent chapter we close with
a transition to the next. We close articles and books with general
conclusions. However, we do handle this shaping in somewhat
different ways than standard, because of the aim of putting the
substantive theory into relief.

Introduction: In writing of introductions, there are several
forms that we do not use. For example, authors often may derive
the problem for the book or paper from a general perspective,
from a literature search or a general interest, or in some
combination of those and with more or less synthesis and
comparative work. However, in introductions we derive the
problem and core variable from the grounded theory, which has
been generated in the research. Existing perspective and
literature are only used as supplements of contrasts, if at all.

Our approach to introducing the problem is to use a “funnel
down” from a “nature discussion” to introducing the problem. The
general, grounded, most relevant properties of the core variable
are discussed, to give the fullest meaning of its general nature.
Then from these properties we select those that will be developed
in the chapter or paper only one of many properties of a core
category. For example, there are several dimensions upon which
clients judge the performance of a professional they visit; cost,
desire to help, kind of help, pace of their service, kind of clientele,
references and so forth. One study focused on the combination of
cost and desire to help. The clients weighted whether the client
thought the professional was most interested either in helping or
in the money. This affected whether or not they returned and
referred the doctor to others.2

To set out the general nature of the core variable and then
funnel it down to a theory on a specific process and problem that
is associated with one property of it is very effective. The general
meaning of the chapter or paper transcends its specificity, thus
putting it in general perspective. Without it the selectivity may
lose general meaning, and seemingly refers to a very limited
study. It starts to appear unit focused. The “nature” paragraphs
may have relevant literature and perspectives woven into them,
as we previously said, but only as supplements or contracts, not
as sources of derivation. The source of these properties, which
establish nature, is their grounding in systematic research.

Once the problem and core variable are “funneled” down to
the purpose of the paper or chapter, it is appropriate to state the
integrative outline established through the sorting. The outline is
written as a cumulative build-up of how the paper will handle the
promise of the purpose. More precisely, the outline discusses each
section and how they are related to each other. Then the reader
knows what he can expect in theory. This promise is fulfillable,
since the analyst is merely stating what he has already generated
and sorted for writing.

If the analyst has not yet codified his outline, or is not sure
of its integration, or indeed finds as he gets into the paper that
the outline falls apart, he should write anyway. He should ask
himself what he should talk about in order to write the most
relevant parts of his theory. Writing can have the consequence of
integrating the outline or reintegrating what has fallen apart. It
is a good way out of a block in integration. If it does not fully
accomplish integration, then rework initial drafts will (discussed
below).

The outline paragraph can be written or rewritten at any
stage in writing. The analyst can do it first or last. It is a matter
of preference. Some analysts prefer, from the beginning, to
establish a tight rein on what they will write. It forces them to
stick to the sorts. Others will do it last when reworking drafts,
after studying what they have done, in terms of their sorts, and
resorts as well as perhaps license to add and subtract yet even
more material. By their writing, analysts are always outgrowing
their previous perspective on the data and some like to leave
options open to change the integration.

Once again, it is a worthwhile exercise for the analyst to
study tables of contents and chapter outlines in published work in
order to develop a grounded perspective on how other authors
resolves this step- if they do resolve it- or forget it or fulfill the
promise of their outline.

When appropriate, a brief methodology of the chapter can be
put in the introduction or relegated to an appendix.

Substantive Sections: The sections, or course, simply follow
from the sorts. They render visible the hard work that the analyst
has done over many months. Thus, they bring the satisfaction
coming from the culmination of the work in a product. In the
analyst’s pent-up demand is too great to deburden himself of his
formulations and to feel the gratification there from, then the
substantive sections or chapters can be written before the
introduction.

Ending the Paper: We have a special view of ending a written
work. First, summaries are not advised. After all, in conceptual
work the paper or chapter is in some manner its own summary.
Students ask us, “How do I finish the paper? I have written the
theory, what else is there to say?” A summary is redundant and
an affront to those readers who have actually read the paper, and
a “cop out” for those who have not read it, however useful to
them. Summaries are usually forced by an editor or brought on by
the analyst who does not know how to end his paper.

Writing a conclusion of recommendations can be worthwhile
if the theory is relevant for practitioners. Our approach to the
ending is to take the core variable, and perhaps a few of those
sub-core variables that worked best, and generate their use and
contribution for formal theory in sociology and for other
substantive realms in sociology. This can be done relatively easily
by brief comparative analysis with data from experience,
knowledge and the literature, and by raising the conceptual level.

Thus, it is easy to see the general import of cultivating in a
study of the cultivating of housewives by milkmen. Since it is a
study of cultivating relationships for family fun and/or recreation
such as in marriage ort friendships? Cultivating can be seen as
occurring up and down social rank: milkmen cultivate up, doctors
often down. Cultivating is a general problem in the service
industries and in the professions. And so it goes: it is not difficult
to bring out such general implications of the core and subcore
variables, which contribute by suggesting other substantive areas
of inquiry to broaden the substantive theory as well as suggest
the important of generating a formal theory. One can also suggest
theory on other aspects of the core variable not dealt with in the
paper, but reviewed in the introductory “nature” paragraph. At
this point the rigors of grounding can be relieved for conceptual
elaborations. We believe that readers find this approach to ending
a paper stimulating, and transcending of the substantive content
given previously.

It must be noted that the generalities of the beginning and
the end sections to the paper are quite different. The beginning
section is systematically generated properties from research
within the substantive area. The end section is generalized
properties applicable to other substantive areas and conceptually
elaborated through non-research comparisons. Substance of time
and place are left behind.

Conceptual Style

One very frequent problem in writing grounded theory is
that analysts have trouble in maintaining the conceptual level
that they have worked so hard to generate. The dictum is to write
conceptually, by making theoretical statements about the
relationship between concepts, rather than writing descriptive
statements about people. Thus, the analyst writes in such a way
as to make explicit the dimensions, properties or other theoretical
codes of his theory as well as the theoretical integration of these
codes.

It is quite easy to slip into excessive description when
illustrating, perhaps because most of us have so much experience
in writing descriptively. So, descriptive writing comes naturally,
conceptual writing does not. It is even easier when the data is
relatively conceptually unanalyzed. The most important thing to
remember is to write about concepts, not people. Thus, one should
write about cultivating or becoming, not milkmen who are
cultivating or nurses who are becoming. (See Chapter 6 on
distinction between units and process.) Saying this is easier than
doing it! If writing momentum is important, then do not worry,
write because the concepts can be brought out during the
reworking stage. Usually initial drafts are a mix of both
conceptual and descriptive levels.

Indicators for the concepts which are descriptive statements
are used only for illustration and imagery. They support the
concept; they are not the story itself. They help introduce the
concept, which can then be carried forward illustration free.
Thus, as we said earlier (in Chapter 5) the dosage mix for
grounded theory is to minimize illustrations, using them for
support purposes, so that the analyst can maximize use of
concepts within the allotted space of the paper or chapter. The
power of the theory resides in concepts, not in description.

The credibility of the theory should be won by its integration,
relevance and workability, not by illustration used as if it were
proof. The assumption of the reader, he should be advised, is that
all concepts are grounded and that this massive grounding effort
could not be shown in a writing. Also that as grounded they are
not proven: they are only suggested. The theory is an integrated
set of hypotheses, not of findings. Proofs are not the point.
Illustrations are only to establish imagery and understanding as
vividly as possible when needed. It is not incumbent upon the
analyst to provide the reader with description or information as
to how each hypothesis was reached. Stating the method in the
beginning or appendix is sufficient, perhaps with an example of
how one went about grounding a code and an hypothesis.

As the analyst learns to maintain a conceptual level, he finds
that it supports itself by becoming more dense and integrated. As
he writes on this level, he should not state in so many words that
he will explain some behavior. He should write the explanation of
how processes actually process problems, so the reader will see
that explanation as such. In short, the analyst should do theory,
not tell that he is going to do it. The latter too easily leads to
excessive in promise, wastes valuable space, and “cops out” by
offering a thin theory. Doing a theory just presents itself as it is:
as modestly dense, integrative and explanatory theory.

Temporal distance from the data helps to maintain a
conceptual level. Sometimes it is best to wait months, even a year
in order to think about the data sufficiently to be able to write
conceptually. Letting sorts or memos lie fallow always helps to
mature the conceptualization of the data. The analyst simply
forgets descriptive details from the field whole his conceptual
scope grows. It is easier to be conceptual sooner in secondary
analysis of other’s data because the analyst never experienced the
field where the data was collected, hence is free of the uncollected
data that lodged in the field worker’s head. 3

There are a few rules that will help those analysts write who
have difficulty in writing. Write as one talks, not as one writes.
This makes writing much easier. So does the idea that if one has
two things to say, say them one at a time. Write the first draft,
with no heed to English construction, so as to focus on the theory
construction. The grammar can be edited later in subsequent
drafts. As with memos, it should not be allowed to interfere with
the ideational out-put. The reader should not underestimate this
problem: many an analyst cannot write because of our concern
with perfect English. Out first concern must be to put over “good”
ideas, which means getting them on paper.

Also, avoid in the substantive sections the use of analogies to
bring out concepts and their relationships. While apparently
useful, under examination any analogy may prove otherwise.
While the current analysis and the analogy (with lots of imagery
such as games, drama or machines) may have a few similar
characteristics, that is often as far as the comparison goes. The
difference in other characteristics between the two undermines
the analysis, unless analyzed straight away. This takes
unnecessary space and time and prevents a straight forward
getting on with the current analysis. For example, in some ways
interaction life may be like a drama, but dramas are very
different than life. This, other properties of drama cannot be
applied to life (such as, “not for keeps”, stage lights, curtains,
directors, etc.). But the catchy drama analogy can take a lively-
minded reader easily down the wrong line or thought as he starts
over-applying drama instead of doing the analysis itself. The
reader is then either lost, not thinking correctly, or is forced to
analyze his way back to the matter at hand, if he cares to.

Reworking

The first draft usually is a delight for the analyst, but also it
usually is very rough. All of its defects can only be corrected by
reworking the draft. As we said, its aim was to capture the
conceptualization and integration of the theory. Like memos, it
was not to be burdened or blocked by the requirements of prefect
English. Until an analyst is an accomplished writer, one half or
more of his creativity typically occurs in reworking his initial
draft.

This reworking may take many trips through the work, as
the analyst solves a problem at a time. Taking on too many
problems at once may prevent doing a good job for each. Writing
is a division of labor process, requiring different jobs of English,
conceptual and scholarly editing. Needless to say, a general
property of the reworking is that as each problem is corrected, the
chances are that it is likely to reveal still other previously
unnoticed problems and possibilities. This phenomenon does
saturate however, or in the alternative the analyst will settle for
less than perfection out of exhaustion and growing personal
saturation.

There are many standard problems for which to rework the
initial draft. They can be seen on two dimensions English and
professional (conceptual and scholarly) editing. The latter
includes weeding out needless redundancy, clarifications or
confused or mixed analysis, trimming and adding illustrations,
footnoting, integrating, reintegrating, weeding out unit focus and
conceptual style and other needs or sections and subsections. We
shall discuss professional editing here with respect to
conceptualization and scholarship. English editing can be hired or
drafted from among friends.

A basic reworking tactic for conceptualization is “flip-
flopping” paragraphs that is making the theoretical statement
come first. Most of us, but beginning writers in particular, often
write paragraphs that start with the description and work up to
the concept and general hypothesis in the last sentence. This
comes naturally and also comes from the constant generating
that goes on. For it to be completely a conceptual writing and to
bring the conceptualization into relief, what is necessary is to put
the last sentence first. Or, “flip-flop” that paragraph by starting
with the concept and then illustrating it through it originally
grew in reverse. Then the concept is imaged, “out front”,
emphasized and usable in carry-forwards. The description is
trimmed to fit the need of illustrating. The same applies to
concepts buried within paragraph if they are the main idea of it.

The carry-forward notion of concepts and the cumulative
build-up of the theory are crucial in reworking. To let a concept
drop may indicate its lack of relevance. And to not have sections
and chapters tied together with theoretical meaning and
development is to undercut grounded theory. All methods we
have detailed previously to this, especially sorting, have set the
writing up for an integrative build-up and the use of relevant
concepts. During the reworking, the analyst makes sure these
two facets or theory generation are there.

In the heat of writing initial draft, it is easy to not tie
sections and chapters together sufficiently. Now the analyst
writes and rewrites these transitions. He makes sure of the
directions of his explanations and bring into relief why and how
each chapter goes in the direction it does. As he reworks he sees
clearly that a concept which has been dropped can be working
usefully in a forward position to enrich the analysis. And if it has
not been used for 100 pages or so perhaps more illustration is
warranted. Missing and messed transitions are easy to spot, with
the perspective of a second or third trip through the writing. This
polishing can be immensely gratifying.

Lastly, it is sometimes useful during reworking to submit
work to colleagues for opinions and critique. If this is too
traumatic, the usefulness is neutralized. The analyst should be
wary and submit to only those colleagues with sensitivity enough
to be appreciative, delicate in suggestion, and knowledgeable
enough to understand and give positive and possible suggestions,
to the reworking.

Submitting drafts to journals is a good source of evaluations
from the outside world or unchosen readers. It is an excellent
source of material for reworking to solve problems that derail the
professional and layman public who do not know the meanings
familiar to and often assumed as general by the grounded
theorist. There is as yet no standardized sociology with respect to
either method of paradigm. This freedom to do different kinds of
sociology is a strength of our field and spawns growth in many
directions. But it also forces accommodations to make grounded
theory accessible to other sociologists with training in different
methods and theorizing. Their critique should be seen in such
light, not as “dumb”, “deprecating”, or “outrageous”.

Footnoting the Literature

One important aspect of reworking drafts is to integrate the
generated theory into the existing literature through the use of
footnotes. The key to this task is the analyst’s attitude toward the
existing literature. His attitude should note be one adumbration,
volume or reverence. It should be one of carefully weaving his
theory into its place in the literature.

To “adumbrate” is for the analyst to find in the literature an
idea he has generated, especially in the literature of a great man.
It is amazing how many authors try to find their best ideas in
previous work in order to legitimate using it, as borrowed or
derived as if they could not be allowed to generate it on their own.
The proper attitude is simply to accept having discovered ideas.
There are so many in grounded theory work! And if the analyst
discovers that one of his many ideas has already been used
elsewhere, the proper attitude is “he (the other author) discovered
it too,” as might any theoretically sensitive analyst in dealing
with the same or similar data. The essential point to remember is
that the discovered idea is relevant because of its connections to
other variables which make up a theory which accounts for
variation in a pattern of behavior. And the analyst will almost
never find this relevance associated with the concept as it was
used previously! Thus, his contribution remains truly original,
since the crucial issue is- a multivariate, grounded theory that
works.

Many a scholar, theorist or empirical research worker will
voluminously footnote ever piece of possibly related literature.
The footnotes seem like a reading list or an extensive
bibliography.4 There are far too many to integrate meaningfully.
Interestingly enough when, in theoretical writings, one studies
these footnotes carefully, one usually discovers that nothing is
referred to that might detract from the originality of the citing
author. This is so even when well known related, relevant works
are overlooked by the theorist, perhaps purposefully, so as not to
threaten his creativity. Thus, much necessary integrated
placement of these theoretical works is missing. This non-
integrative approach cannot fail to hinder the growth of theory.

Reverential, commemorative, and referral footnotes are fine,
as long as they do not take precedence over the generated theory.
They go hand and hand with integrative placement of the
grounded theory. There is no magic about a theory in print before
the analyst’s writing just because it already occurred that
warrants undue reverence. Soon the analyst also, will be in print
and his ideas will be used. Thus, reverence and commemoration
should be moderate based on what the idea from the literature
truly contributes to the big picture, just as the analyst uses ideas
for his own theory. Idolization of “great men” should be replaced
with the attitude “He too was working with these ideas.” In
addition, there should be implication that the current idea was
derived from a previous author’s, merely to legitimate the idea. In
our research ideas are discovered on their own or emergently fit.
Clearly reverential derivations are farthest from our
methodological position.

These efforts should be as short as possible so as not to
derail a reader who stops to see the footnote. A reader will also be
less likely to miss footnotes, because they are brief, since he can
see at a glance that his reading will barely be slowed. Footnotes
that require length can be put at the end of the chapter as
annotated references. Even longer requirements can result in
another article.

Obviously this kind of footnoting takes analytical work: it is
not easy. But it is done just as the analyst does his grounded
theory: he compares, generates, memos, sorts and writes-up the
ideas for the footnote.

Theoretical Pacing

It is appropriate to close this chapter by referring to many of
the properties detailed in Chapter 2 [of Theoretical
Sensitivity] on theoretical pacing as the apply in writing. The
theoretical pacing of reading, talk, deadlines, respites,
collaboration and personal growth, become very relevant during
writing.

Reading: We have said that during data collection, coding,
memos, and sorting of memos, the analyst should read in other
fields so as not to preempt his thought regarding the significant
variables in the substantive area under research. The analyst
should continue this rule throughout the initial draft, if his
sorting has not reached a firm integration. This maximizes on
another dimension the emergence of his theory.

But when he starts reworking his draft he should make a
concerted effort to cover as much literature as possible in the
same area in which he is writing his theory. Now the job is to
compare his work to others and weave it into its place in the
pertinent theoretical and substantive literature. It also sensitizes
the analyst to reworking his theory to the best advantage, as he
studies how others are theorizing in the field. As noted above,
integrative placement of ideas by supplementing, extending, and
transcending other’ work is the issue, not their preemption of his
ideas.

It is a travesty not to do this scholarly aspect of grounded
theory for sociology, though some analysts do not because of their
personal saturation. Just because grounded theory has emerged
and can stand on its own, does not mean it should be left to
isolation or only for the consumption of laymen interested in the
area. It should contribute more explicitly to the “bigger
enterprise” in some way. If theoretical and substantive literature
is sparse, as it has been for some of our own studies, hopefully it
starts a literature to which others can contribute.

Talk: As in doing codes and memos, the analyst should avoid
talking about the ideas he is writing. At best, talk is interrupting
and distractive. At worst it gives away ideas before writing by
releasing the energy behind them which can easily be followed by
forgetting them or feeling no need to write them up. Also others
can derail or block even the most careful writing up of sorts. Once
the analyst is deep in the writing mode, he should stay there
undistracted. There is plenty of time during reworking to discuss
ideas for critique, clarification and polishing after the initial
draft. At this point they are down on paper so they cannot get lost
or blocked. The initial draft can always be changed, it if is
written. But we have seen too many drafts get blocked or
prematurely changed or closed off, by a too soon critique of ideas
by a trusted colleague, who has little notion of the interrupting
effect of his ideas through connections to other codes that he is
unaware of.

Collaboration: A carefully applied exception to the rule on
talk is to seminar with a collaborator who is stimulating rather
than draining. Again, when writing, the analyst must be careful
because of possible blockage, derailment, and/or drain from even
this trusted, respected source. There is really no reason why
collaborators cannot also wait to talk during reworking, once they
know which parts of the integrative outline they will write up.

Collaboration is very useful in reworking, because it saves
much time. An analyst may have to wait a month or two, to be
able to rework his draft with sufficient freshness. While a
collaborator can start reworking it the next day, since for him,
the initial draft is fresh (not having written it). When
collaborators trust each other with reworking of their final drafts,
then writing proceeds very fast. When they do not trust, they can
destroy each other. 5

Collaborators who are out of “sync” with each other’s pacing
should be patient in waiting for the other to be ready to talk.
Demanding talk can be damaging to the work and the
collaboration. It may force premature closure of the writing of one
collaborator, when the other’s judgement is valued.

By the same token to demand talk of a personally saturated
colleague who can not say one more work about the project is to
be avoided. At this point the collaboration is either over for the
moment or completely.

Deadline: Our goal in preventing talk and showing one’s
work before the initial draft is to maximize the energy behind
productivity and minimize those circumstances which so often
short circuit it. Helpful along these lines is the analytic rule of
giving oneself the shortest possible deadline for the initial draft.
This pressure prevents wasting time on premature showing and
talk. And it gives the analyst an expectation to himself and others
as to when he can show his work. A deadline is strength inducing
to ward off these and other typical foibles of writing. It prevents
drift, evasion and over elaboration of the theory. It generates
focus, perseverance and closure.

A deadline should include the possibility of respites
consistent with the analyst’s personal pacing recipe. Otherwise
the work may become a drudge that undercuts the richness of the
writing. The deadline and respites should be synchronized both
with the analyst’s personal pacing and the natural pace of the
work. Respites occur best after semi-closures, such a finishing a
section or sub-section.

Outgrowing the material: From the outset grounded theory
work is a growing experience both personally and with theoretical
understanding of the data. Writing further grows the analyst
with respect to maturity with his data, and fortunately, knowing
far more than he is capable of getting on paper. The sheer fact of
writing a paragraph, quite often, yields insights that put the
analyst beyond it. This outgrowing of one’s material can be
disconcerting and even undermining of the final writing of the
theory. In grounded theory work the analyst must realize that
writing is but a slice or a growing theory.

The analyst, who feels that he cannot finish writing because
he can never begin to tell what he knows, should just accept this
fact and finish as sorted and planned. He can never outstrip his
own constant growing, no matter how much he writes. His
writing will always spawn growth and yield more to say. He
cannot overload his work and break his integrative outline- thus,
he must accept that although he knows more and better, his
reader, knowing less, can greatly benefit by whatever the analyst
does write. It will be “news” to the reader, even if “old hat” to the
analyst. Others will respond to the richness of the dense
grounded theory, while the analyst may feel he had only begun
and that it is “sort of thin.”

It is a tribute to grounded theory that it maximizes this
outgrowing of one’s theoretical material. The reduction, natural
high and relief from closure on what theory he has written,
usually outweighs the nagging realization, that much more could
be said. Yet some analysts still are blocked by the “puniness” of
writing compared to what they really “could tell.”

Other qualitative methods leave much theory implicit and
underdeveloped, because they do not allow for much generating,
strategies of coding, sorting, memoing and integrating. These
likely will leave the theoretically inclined researcher with an even
worse feeling that much has been undone and left out, since he
has not at least integrated a fledgling theory that fits and works.

The point is to publish this “slice” of a growing theory so
others can get to this point and also use it and grow with the
theory. The differential perceptions of the reader and the writer
do not redound against the writer. He will be applauded for what
he did, not what he knows he did not do. What he did not have
room to set down can be covered in other papers or books and can
be suggested to others as future research leads. What is arbitrary
about writing and publishing a substantive theory is more than
compensated for the contribution of the grounded theory
methodology by which the theory was generated.

1 Originally published as Chapter 8 in Glaser (1978). Theoretical Sensitivity, Mill
Valley, CA: Sociology Press, pp. 128-141.

2 David Hayes-Baptista: This paper was a Master’s thesis at University of
California, San Francisco, Sociology Program.

3 Barney G. Glaser, “The Use of Secondary Analysis by the Independent
Researcher,” The American Behavioral Scientist (1963), p.p. 11-14.

4 Robert K. Merton: Social Theory and Social Structure (Glencoe, Free Press, 1949)
and Neil Sidser, Collective Behavior (New York Free Press of Glencoe, 1963).

5 Warren O. Hagstrom, The Scientific Community (New York: Basic Books,
1965) Chapters 11 through 15.