Computer-induced changes in intellectual and scientific work
A sociological perspective
(14 May 1996)
By taking a closer look at the characteristics common to all processes of text production, it is found that producers exploit three fundamentally different sources of information:
While modern artistic (literary) production is characterized by high reliance on intrapersonal "subjectivity", scientific work is supposed to be guided by interpersonal communication networks within professional disciplines as well as by systems of worldwide institutional publication.
In legal thinking, institutional sources (like governmental laws or court decisions) are supposed to prevail - while interactive communication processes and purely subjective judgements are thought to occupy only a residual place.
In the following, it is argued that the computer generates the basis for fundamental changes in the way these three axes of intellectual activity are functioning - as well as in the way they are interrelated to each other.
In all three dimensions, the functional significance of computer-supported information-handling is based on three pillows:
Focussing on the realm of text production, its outstanding performance is evidently that of an "externalized personal memory": giving the individual autonomous control of the access to the whole complexity of his (her) past written products for browsing, reordering, recombination, modification or deletion.
Thus, the notion of "subjectivity" has to be redefined in
order to include personal computer-stored documents operating as a base
for self-referential mental activities. Like personal diaries or
confidential letters, they belong to a sphere of "privacy"
needing moral acknowledgment as well as legal protection.
1) ExternalityTraditionally, any written text was likely to be expressive of the author's internal state of mind at the very moment of its production (or last revision). Even when writers relied on notes or draft versions, these writings were mere "half-way products" needing refinement and modification in the light of actual considerations.
The availability of externalized computer memories makes it more likely
that text producers rely directly on past writings of their own, while
reliance on actual subjective factors (evoked subjective memories,
emotions, accidental information etc.) declines.
From the perspective of the philological sciences, this means that it
may become impossible to reconstruct the "intellectual history"
of a monography or the whole life work of an individual thinker, because
the available texts contain no hints about the time periods when certain
passages were created or revised.
Externality may also facilitate reflective acts of self-evaluation and
self-critique. By editing older documents, individuals meet their own
"past selves" in a far more explicit and objectified sense than
by simply "remembering" previous thoughts and action. As a
consequence, they are better able to assess how much and in what way their
thinking has changed in the meantime or to gain distance to their present
self by re-identifying with earlier points of view.
2) ComplexityAverage hard disks have enough space to store all texts ever produced by even the most prolific writer. In contrast to the physical memory, all components are maintained at the same high level of accessibility and explicit precision, regardless of their number and size and irrespective of the time span since they were created.
As a consequence, computers enable individuals to create more complex and more differentiated "private intellectual worlds" potentially containing at any moment the accumulated stock of all earlier writings.
The richer this self-generated intellectual environment, the more writers may become disposed to become absorbed by self-referential activities while reducing their attention to what other people have to say.
In other words: whenever they look for suggestions or information while writing, they are more inclined to exploit their own accumulated memory than to consult foreign texts or communicate with collegues.
In intellectual fields where high self-referential thinking has traditionally been the rule (e.g. in philosophy), computers may therefore reinforce the trend that individual writers cumulatively drift away from each other by submerging in a self-made, idiosyncratic intellectual world where they exclusively meet their different past selves explicitly incorporated in the documents stemming from different phases of their own life.
Wherever the cultural environment emphasizes values of individualism,
personal computers easily lend themselves as potent catalyzers of
interindividual differentiation: by adding to the subjective world of
psycological consciousness a complex second world of "externalized
privacy" where self-referential activities can take place.
3) ControllabilitySelf-control over the psychological memory system is rather limited because thoughts follow each other in a rather spontaneous, associative fashion and because most stored items can be evoked only imprecisely and suffer from obsolescence during time.
Vis-à-vis their externalized computer memory, individuals find themselves in a far more autonomous position, because they can switch it on and off ad libitum and can easily evoke every item they wish with full precision at every moment of time.
As no principles of "forgetting" are operative in order to reduce the available complexity, individuals are even forced to make their own selections about what should be evoked or suppressed at any one moment.
In order to organize their externalized intellectual world, they have to apply their own criteria of classification and categorization.
For instance, in the way they order documents in different folders or in the names given to different documents, individuals convey their personal conceptions of what belongs together and what should be kept apart.
Thus, computers catalyze the genesis of individualized systems of ordering information and products of intellectual work.
Compared with traditional ordering devices on the basis of written cards or sheets, these new systems are far more flexible (and reversible) because computerized search procedures allow that multidimensional systems are applied and that modified or additional selection criteria can be introduced at any point in time.
Another consequence of high controllability is that individuals are
better able to standardize and routinize verbal expression habits or
explicit thought structures by re-using the same formulations or text
components on very different occasions.
1) ExternalityMany important functional consequences of computer-supported communication for intellectual activities result from the basic fact that communication flows are extremely dissociated from all physical, behavioral and psycho-social contingencies of individual emitters and receivers.
The whole meaning of E-messages has to be inferred exclusively from their intrinsic verbal characteristics, because absolutely no nonverbal cues (like tone of voice, mimic or gestic movements etc.) are provided that would help to specify their interpretation, evaluate their credibility or to judge how much importance is attached to them by the senders.
This handicap is even more acute than in the case of conventional
letters because status characteristics of interaction partners.(gender,
grade, age etc.) or even full names are often not known.
Of course, the advantages of status levelling accrue particularly to incumbents of low status positions (while high ranking individuals may correlatively lose because they are no longer able to display their superior prestige). This explains the empirical finding that more electronic contacts are directed from lower to higher levels than the other way round. (Welsh 1982).
As nonverbal evaluation cues are lacking, assertions have to be believed and arguments to be honored for their own sake: not because they originate from an authoritative source and not because rhetoric abilities are displayed to make them sound "convincing".
It is widely acknowledged that the dissociation of verbal messages from their source of origin has been a major correlate of all writing since its inception 5000 years ago (see: Bolter 1984:133/134), and that has been a precondition for evaluating verbal statements on the basis of objective criteria like logical consistency or factual truth.
If this is granted, it has to be added that computer technology creates an additional platform for the development of a completely de-subjectified intellectual culture: by extending this dissociation (of messages from senders ) from the narrow and static realm of written documents to the wider and more dynamic field of interpersonal communication.
Of course, E-communication is most adequate in cases where words have a
highly standardized and consensual meaning, so that there is no need to
provide additional specifying cues. Thus, it will be used widely in
scientific fields characterized by high paradigmatic development and
formalized codes of expression (mathematics, chemistry etc.) and less in
most social sciences or humanities, where imprecise definitions and
nonconsensual terminological interpretations prevail.
2) ComplexityBy combining the informality and interactive flexibility of telephone conversations with the high transmission capacity and precision of mailed documents, computer networks can carry communication processes higher in variety and complexity than any of the conventional translocal technology alone.
Thus, it is possible to contact many different interaction partners very quickly (or almost simultaneously in multilateral arrangements), to engage in regular or even continuous processes of interaction and to transport very complex packages of information (e.g. integral data sets or manuscripts) within a very short time.
As a consequence, intellectuals embedded in computer networks are
likely to draw their knowledge or ideas from a larger absolute number of
other individuals (Huber 1990; Foster/Flynn 1984) and to confront
themselves with a more heterogenous variety of different points of view.
In general, the capacity of electronic communication systems to
increase complexity and divergence seems to be much higher than their
ability to reduce complexity by generating consensus and eliminating
deviant opinions. A major reason for this is the lack of nonverbal levels
of communication (e.g. by displaying status differentials or superior
rhetoric abilities) that makes it difficult to exert leadership or to
implement efficient strategies of persuasion: "Groups take longer
to reach agreement and consensus is less likely. Controlled experiments ..found
that, compared with face-to-face groups, computer conferencing groups
needed more time to reach a decision and were less likely to reach a
unanimous decision for complex problems." (Kerr/Hiltz 1982: 145).
As a consequence, an additional need for consensus-producing face-to-face interaction may be felt in order to compensate for the centrifugal effects of electronic interaction.
By making factual dissensus more visible, computer-supported
communication may well contribute to the decay of broader "scientific
communities": by proving that their "unity" and "cohesion"
has been founded on the illusion of sharing common paradigmatic views.
Instead, a more pluralistic co-existence of smaller infra-disciplinary
subcultures is likely to emerge: each centering on its own networks of
Countervailing the growing complexity of translocal
communication, the complexity of local communication systems can be
decreased. For instance, it is no longer necessary to incorporate into a
residential research team all the kinds of qualifications ever required,
because any translocal informants can be contacted whenever a need for
highly specific information or advice arises.
As a consequence, research organizations and research groups can remain smaller and larger percentages of researchers may be able to work completely "alone". This again may increase the flexibility of research units and their capacity to reach quick and consensual decisions (Huber 1990).
When far-away interaction partners are accessible on a continuous basis,
it becomes more decisive to be at least superficially acquainted with the
"right people" all over the world. Therefore, the relationships
initiated at international workshops and conferences are becoming crucial
because, by electronic communication, they can be an important factor in
everyday intellectual work.
Finally, we may also see the emergence of extremely specialized new
professionals that offer their outstanding, but narrowly focussed
knowledge worldwide through electronic channels.
3) ControllabilityConventional forms of communicative interaction imply considerable restrictions of autonomy, because individuals have to be present at certain places for specific periods of time (conferences) or have at least to synchronize their communicative engagement and separate it sharply from other activities (telephone).
Computer networks give rise to a new type of communicative participation role, outstanding for its almost complete lack of interference with all other kinds of individual roles and activities.
Irrespective of their geographical location and many other situational contingencies, individuals can chose any time to send any messages to anybody - as well as deciding themselves when to read incoming messages.
The almost limitless compatibility of electronic communication behavior with various personal characteristics, roles and situational conditions is also anchored in the fact that anonymity encourages a highly informal communicative behavior contrasting sharply with the ritualistic formality of written letters:
"...people using electronic mail overstep conventional time boundaries dividing office and home, they mix work and personal communications, they use language appropriate for boardrooms and ballfields interchangeably; and they disregard normal conventions of privacy (for instance, by posting personal messages to general bulletin boards). This behavior is not counteracted by established conventions or etiquette for computer communication. There are few shared standards for salutations, for structuring formal vs. informal messages, or for adapting content to achieve both impact and politeness." (Kiesler/Siegel/McGuire 1988).
As a consequence, prevalence, intensity and kind of communicative participation are highly dependent variables: determined almost completely by subjective needs and motivations or situational antecedent factors. (Johansen/DeGrassse/Wilson 1978).
This contrasts most sharply with face-to-face conferences where participants are drawn unwillingly into ongoing communication processes as a result of their mere physical presence.
While controlling their own communicative behavior, electronic
interaction partners are - because of this same fact - extremely unable to
influence each other. Therefore, computer conferences often suffer from
"lack of discipline" caused by unpredictable changes in the
intensity and quality of communicative dispositions and motivations (Kerr/Hiltz
1) ExternalityExternalized storage is the common characteristic of all written information. But in the traditional world of published books and journals, externalization remains incomplete because texts are still primarily segmented according to individual authorship: and this "genetic" segmentation heavily determines the ways of diffusion and reading.
Thus, intellectual culture is dominated by author-orientated strategies of digestion, because:
- it is very easy to have access to everything the same author has written (by reading his/her book of "collected works")
- it is difficult to collect what very different authors have said about the same topic (because many different books have to be aggregated and worked through in a time-consuming manner).
This "author-bias" again gives rise to systems of intellectual or scientific reputation where whole individuals are the targets of prestige attribution - an important basis for motivating intellectual producers and for structuring intellectual careers. (Luhmann 1968: 147-170).
While computerized libraries still order their entries according to
authorship, this production-oriented criterion of text-segmentation
is likely to get de-emphasized or neutralized by cross-cutting reception-oriented
selection criteria chosen by the different users.
As a consequence, the probabilities of being noticed in the intellectual world are decisively reshuffled. Even very modest authors can easily get attention when they have expressed something exactly to the point, while most prestigious scholars may be ignored without mercy when nothing relevant is found in their contributions.
As author-oriented criteria of selection (e.g. reputation, excellence, etc.) decline in functional importance, intellectuals and scientists may become faceless, anonymous contributors who have to be satisfied to be widely "used" without hope that even quite extensive "usage" adds up to a considerable individual reputation.
This is most vividly seen in the case of expert systems which
belong to the most collectivized products of human intellectuality,
because originators of specific knowledge items and decision rules can no
longer be identified at all.
2) ComplexityIn traditional intellectual work, the richness of available information was heavily reduced by "natural" obstacles of accessibility: e.g. because exotic journals or the books of "unknown authors" were not readily at hand.
As a consequence, systems of intellectual and scientific publications were quite "imperfect markets": characterized by authors, "schools" or national subcultures having "partial monopolies" within specific institutions or geographic regions (Malone/Yates/Benjamin 1988).
In computerized libraries, the most famous and the most marginal documents are retrievable on the same plane of accessibility, so that search procedures are likely to produce very hetergeneous materials from all kinds of (even quite unexpected) sources and locations (Malone/Yates/Benjamin 1988). Therefore, markets of intellectual exchange become more "unbiased" because local "citation subcultures" are likely to vanish in favor of more homogeneous cosmopolitan cultures (Malone/Yates/Benjamin 1988).
Thus, scientists are likely to find that the "paradigmatic
consensus" they counted on does not really exist, and judges and
lawyers using encompassing legal data banks may well become quite confused
because when asking a specific question, they encounter not only official
statutes or highest court decisions, but also additional (otherwise
ignored) sources of "valid law" (e.g. older sources of
non-codifed rules and practices, academic opinions etc.) (Endrös
1987:110; Raden 1989).
Wherever a single specific solution has to be found (like in legal matters), more explicit (=consciously chosen) selection criteria have to be applied in order to rule out alternatives, because implicit mechanisms of complexity reduction (like physical difficulties in accessibility) are no longer operative. In the legal field, this means for instance that additional criteria for defining priorities among different sources of law have to be applied.
Electronic data banks may stifle innovative problem-solving insofar as
they make it more likely that a past solution to any actual problem
3) ControllabilityThe growing autonomy of readers is a consistent trend of cultural evolution. While papyrus rolls enforce a rigid following of successive lines, books and newspapers can be opened at any chosen page (Bolter 1984:137;161).
But print media still tend to "partialize" the world of written culture along the lines of production units, because opening one book (or journal) means being non-attentive to all the others.
In addition, even encyclopaedias or dictionaries determine reading behavior by fixing immutable criteria of classification (e.g. by alphabetical order).
Computer-supported storage systems provide at least the potential to
skip through immense text material for any purposes in a very short time,
so that strategies of decoding are no longer determined by encoding
structures at all.
This fact has three most important implications:
As a consequence, information transfer between spheres was quite difficult and highly selective because burdensome transcoding activities (e.g. writing up thoughts, protocolling discussions, presenting papers at conferences etc.) had to be undertaken.
By contrast, each personal computer can function as a device where these three worlds can meet, interpenetrate or even merge with a minimum of effort according to the preferences of any individual user.
For instance, "private notes" on hard disks can easily be
turned into communication messages without changing form and content (or
the other way round), while excerpts received from data banks can be
assimilated into the externalized private memory or fed into computer
conferences without any recoding effort or loss of information.
Extremely "collectivistic" new procedures of text production are possible: for instance when several individuals located at different places work cooperatively on the same text, so that no specific text components can be attributed to any specific authors. (Kiesler/Siegel/McGuire 1988:658).
An important consequence of this is that, in comparison with face-to-face interaction, communicative processes are less dominated by ideas which participants have actually in their minds or readily "at hand", because they are able to edit and transport stored text components created anytime in the past. Therefore, "replies" can reach higher levels of complexity and sophistication than any oral discussions.(Kerr/Hiltz 1982: 126).
It is also possible to make institutional data banks "interactive"
by allowing users to comment on a paper and letting authors reply. (Folk
1977:78f.). All subsequent users would then also receive all previous
discussion statements - and may well feel motivated to add their own.
In the case of electronic information technologies, these mediating impacts are particularly strong because computers are the most universal of all human instruments: lending themselves to support an unlimited multitude of different individual activities, social interactions and cultural patterns. In accordance with this very general proposition, it has been argued above that computers are equally apt to reinforce self-centered individualism as well as networks of social communications and structures of institutionalized tradition.
As a consequence, adequate predictions about the future impact of
computer technology are only possible in spheres where reliable and stable
co-determining conditions (on the individual, social or cultural level)
can be identified.
But in the field of intellectual activities, individual idiosyncrasies
and informal interactional relationships are so prominent that it is hard
to identify the social antecedents that will determine to what extent and
in which way electronic media will be applied in the future.
On the other hand, we see thousands of writers, journalists and researchers in the process of getting acquainted with personal computers as a tool for supporting and organizing their own work, for contacting distant interaction partners and for investigating data banks and other depositories of information.
But even these communicative and institutional applications are likely to remain on a low level of institutionalization, because in modern (and a fortiori in "postmodern") societies, individualistic behavior patterns are supported not only by cultural values , but also by the growing pluralism and objective divergences of ideas, paradigms and methodological approaches.
As a consequence, the impact of computers on intellectual work (as well as on work or human behavior in general) will never be adequately assessed by means of empirical research because empirical studies at best mirror the actual (and more frequently:
past) stages of applications.
Thus, computer technology offers a starting point for a new kind of
(meta-)intellectual activity: imagining various possible development
patterns ("scenarios") for alternative future intellectual
Endrös, Alfred. (1987), Rechtssprechung und Computer in den neunziger Jahren , Berlin, Springer Verlag.
Folk, Hugh. (1977), "The Impact of Computers on Book and Journal Publication" in: Divilbiss, J.L. The Economics of Library Automation , Urbana Ill.
Foster, L.W., Flynn, D. M. (1984), "Management Information technology. Its effects on organizational form and function", Management Information Systems Quarterly, Vol. 8.
Huber, George P. (1990), "A Theory of the Effects of Advanced Information Technologies on Organizational Design, Intelligence, and Decision Making" in: Fulk, Janet, Steinfield Charles, (Eds.) Organizations and Communication Technology , Newbury Park, Sage Publications.
Johansen, R., DeGrasse R. jr., Wilson, T. (1978), Group Communications through Computers, Vol. 5, Menlo Park, California, Institute for the Future.
Kerr, Elaine. B., Hiltz, Starr Roxanne. (1982), Computer-Mediated Communication Systems. Status and Evaluation ., New York/London, Academic Press.
Kiesler, Sara, Siegel, Jane, McGuire, Timothy. (1988), "Social Psychological Aspects of Computer-mediated Communication" in: Greif, Irene. Computer-Supported Cooperative Work: A Book of Readings , San Mateo, Cal., Morgan Kaufmann Publishers Inc.
Luhmann, N. (1968), "Selbststeuerung der Wissenschaft", Jahrbuch für Sozialwissenschaft .
Malone, Thomas W., Yates, Joanne, Benjamin, Robert I. (1988), "Electronic Markets and Electronic Hierarchies" in: Greif Irene. Computer-Supported Cooperative Work: A Book of Readings , San Mateo, Cal., Morgan Kaufmann Publishers Inc.
Martino, J.P. (1972), Technological Forecasting for decision making , New York, Elsevier.
Strassman, P. (1985), "Conversation with Paul Strassman", Organizational Dynamics , Vol. 14, (2).
Welsh, L.A. (1982), Using electronic mail as a teaching tool. Communications of the ACM , 23.
Hans Geser: Home Hans Geser: Online Publications