CHAPTER 6

STUDY RESULTS

Introduction

While the analysis within this study focuses on DA of a message corpus (Fairclough 1992, Yates 1994), contextual information, as a form of triangulation is also necessary, to provide the reader with evidence from which to judge the confirmability of the report. The following sections, present and discuss data on list usage, together with findings from a small number of interviews conducted by e-mail with list subscribers. Detail of the analytic method within each of Fairclough's three dimensions is provided as aspects of the corpus are discussed.
 

Data about the list and usage

Some other nursing lists are moderated, i.e. all messages are approved before redistribution to subscribers. Both approaches have advantages and problems; moderation may be viewed as censorship, although it is generally used to restrict obscenities, control 'flaming' and ensure that messages posted are relevant to the list's purpose. The potential for such problems on unmoderated lists is great, but experience of using the Nursenet list indicates that the frequency of the first is negligible, and of the second low. It is questionable whether anything that might conceivably impinge on nursing and health care could be considered irrelevant.

When the corpus was selected, the number of non-concealed subscribers to the list was 1237 (Norris 1994a), representing 20 countries. The overwhelming majority (85%) were from North America (USA 62%, Canada 23%). 4.5% were from European countries (including under 1% from UK), with the rest from Australia (7%), New Zealand, southern Africa, east Asia and South America (see Table 6.1).

Table 6.1. Nursenet subscribers by country as at 31 October 1994 (Norris 1994a)
 

Country	Number of subscribers	Percentage
USA	763	61.7
Canada	287	23.2
Australia	91	7.4
GB	12	1.0
Europe (exc. GB)*	44	3.6
New Zealand	11	0.9
South Africa	4	0.3
Other**	25	2.0

* Belgium, Denmark, Finland, Germany, Ireland, Italy, Netherlands, Norway, Sweden represented

** Columbia, Japan, Thailand, Zimbabwe and non-ascertainable

As is commonly found, due to the absence of physical links to telecommunication networks, large parts of Africa, Asia and South America are underrepresented.

Males make up a larger proportion of Nursenet subscribers (about 21% (Norris 1995a)) than of the nursing population generally, estimated at 5-10% in 'developed' countries. There are variations between countries, and analysis of subscriber names reveals that, for example, 25% of Australian subscribers are male, 10% from Canada, 58% of GB and 17% of USA subscribers. These are probable underestimates, as I made no assumptions as to gender where names were equivocal, or just initials were given. The list, with 80% of subscribers female, is substantially different from many other discussion areas on the Internet (where users are estimated to be 95% male (Grant 1994)); this is likely to influence the nature of the discussions. This gender difference also makes problematic any attempt to generalise findings from this study to male-dominated discussion areas.

The number of messages posted to the list per day varies. It was monitored for a 3-month period including the dates used for corpus selection. The mean number of messages posted per day was 21.5, with a range of 2 to 50. Most frequently, 11-25 messages per day are posted, although the average number rose to 33 around the time of the corpus. This message frequency does not necessarily mean that this many people posted messages; frequently a subscriber may post several messages in a day, following the netiquette convention of confining messages to a single topic.

Subscribers, by default, receive messages as each is posted, but may choose to have them collected and sent once a day as a 'digest'. If cost is a factor, i.e. the subscriber has to pay a charge for each message received, they may prefer the latter option. About 26% of subscribers received digests (Norris 1994x), and receiving messages in this form would be expected to have an impact on the way in which subscribers can respond to messages (see Chapter 6).

As with most areas of the Internet, the list's lingua franca is English (or American). Participation is therefore restricted primarily to those who can read/write English, wish to participate, know of the existence of lists, and have the necessary computer literacy and access (Aycock & Buchignani 1995). In respect of this list, nurses from many areas of the world are precluded from discussion, although not due to any wilful act on the part of the list-owner or subscribers. There is nothing, provided the technology is present, to preclude other languages being used in discussions.
 

Interviews with subscribers

After an e-mail message was posted to the list inviting interviewees, nine participants (note 3) provided some information relevant to the study, although detailed interviews were conducted with only five. Most of the participants (5 of 9) were in education, either as academics or as postgraduate students; two were senior nurses in management roles, and the final two worked in miscellaneous roles that did not primarily involve direct clinical contact. This finding, that none of my participants had a significant role in direct patient contact, is similar to the impression one gets from the discussions on the list generally. Some frequent subscribers to the list, it must be noted, do seem to have a significant clinical element to their work.

While all of the participants use the Internet for CMC, their purposes vary somewhat, although the impression given by most of their primary purpose is illustrated by figure 6.1.

Figure 6.1. Text sample from interview with Nursenet subscriber.

My primary reasons for using computer mediated communications, as they are applied to nursing, are to keep myself in touch with the current state of the nursing profession and to keep myself up to date with new and possibly more appropriate techniques than the ones I currently employ in my nursing practise. The aim is to provide any patient, within my care, with the best healthcare service that I can (participant 5).

Email is a well accepted communication tool in most institutions and has been for some time (part. 9).

The nurses who use email include the staff nurse at the bedside, the nurse managers, ...In addition, the Clinical Teachers and Clinical Nurse Specialists all use email as their primary communication tool (part. 9)

Nursing staff use the computer e-mail primarily for clinical communication. Generally not educational, except for brief paragraphs about new clinical approaches, equipment, or problems. ... For instance, the decisions of patient care conferences are sent out on e-mail, updates on patient conditions are shared, new problems are identified, etc. (part. 1)

our ward terminals to be connected to our library computersystem to allow immediate access to on-line databases, such as the National Library of Medicine. Then clinical staff could have immediate access to clinical information at any hour of the day (part 1)

In view of the literature on the subject, I felt it important to ascertain participants' perceptions of the personal or impersonal nature of CMC, whether they saw it as speech or writing, and the situation of self within the communication. Most participants view CMC as neither strictly speech nor writing, and some talked of a new hybrid of the two, with a spontaneous nature, akin to speech and less formal than written communication (as in Yates' (1994) findings). They also recognised the aporia implicit in viewing as speech a form of communication that is both produced by the author and consumed by the recipient in written form (figure 6.3).

Figure 6.3. Text samples from interviews with Nursenet subscribers.

a]

This communication actually seems like a hybrid to me of speech and writing. Certainly less formal than normal written communication, but physically in written form. ..When I communicate via e-mail in this manner, it is fairly spontaneous, much as face-to-face speech would be (part.1).

I think that some view it as conversation. Many do not bother to edit what they write, use horrible grammar and spelling ... Others are more careful, and seem to understand that this is a written form of communication. I suspect that many of these may be beginners. As you become more comfortable, you easily adopt the informality of this method of communication. My personal belief is that this method of communication is neiter spoken nor written. It can be preserved for the future, but is often off the top of one's head. It is a new method of communication...I often wonder if any of us really understand this. (part. 4).

Most of the participants viewed their communications via this medium as very personal and personalised (figure 6.4). Whether this is a peculiarity of nurses, or a feature that is developing more generally on the Internet with greater use is an issue that would merit further study.

Figure 6.4. Text samples from interviews with Nursenet subscribers.

a]

I find communication via computer to be rather more personal than impersonal. This has been an interesting discovery, and something I would not have expected without hands-on experience ...I think I get a rather clear picture of the other person's personality via the written word. (part. 1)

I have not read the literature, but I know that this contact with you is personal. I've never had a book address me by my name nor have I had a book answer my questions as I frame them (part. 5)

Humour can fail horribly, people don't know when you are serious or not, and the lack of personal contact makes "flame wars" very easy to start. (part. 3)

The corpus and its context

Two days were selected in advance, which were weekdays within the academic year, when network traffic is higher than during weekends and holidays, and the total messages posted to the list were downloaded, although in digest form. There were 29 messages on the first day, and 41 on the second.

It is useful to discuss characteristics of the nurses posting the messages. Of the 70 messages forming the corpus, at least 22 (31%) were posted by males (the figure may be higher, as no assumptions were made where gender was ambivalent), and of these 22, 12 were from the same person. This probably reflects netiquette conventions, rather than any other explanation, although no female contributor posted more than 4 messages to this corpus. This is similar to the use of the list seen over a longer, randomly selected, period of 14 days, during which 163 subscribers posted a total of 263 messages, with most (70%) posting only 1 message, and only 24 (15%) posting 3 or more messages. Of those posting the highest numbers of messages (6 or more each), 3 of the 4 were male, with again a male posting the highest total number (12). This is not dissimilar to Kaye's (1991:11) findings that "read/write ratios (number of messages read:number posted) typically are of the order of 100:1...and a relatively few percent of active users often account for up to 80% of message traffic". The congruity of these findings with the suggestion from the literature of the equalising potential of CMC is addressed later.

At least 54% of the messages emanated from the USA (note 4), with another 28% from Canada; thus only 17% of messages to this corpus originated outside North America, a figure similar to the number of subscribers outside this area. At least 50% of the messages originated from subscribers with academic or educational e-mail addresses reflecting the historical development of the Internet within the academic community.

Following an initial analysis of the corpus, and identification of several genres, it became apparent that a close analysis of all the messages posted, including their intertextual links within and outside the corpus, would be impractical within this work. Fairclough (1992) suggests that examination of what he calls "cruces", i.e. "moments of crisis" in the discourse can provide a useful point of inquiry. Such aspects of the discourse, by providing opportunities for participants to themselves closely examine and correct potential misunderstandings through repair, repetition, clarification, etc. highlight "aspects of practice which might normally be naturalized, and therefore difficult to notice; but they also show change in process, the actual ways in which people deal with the problematization of practices" Fairclough (1992:230). The corpus included just such a crux in the discussion of what might be summarised as "who should define nursing?" While the following pages contain some discussion of aspects of the whole corpus, the main focus of the analysis will be on the 21 messages that form a discussion around this crux.
 

Analysis of the corpus

Discourse practice

Discursive practice refers to the processes by which texts are produced, distributed and consumed, which Fairclough (1992) asserts vary between discourses and according to social factors, including norms and conventions. While discursive practices can be seen as reinforcing conventional modes of discourse, and "contribute...to reproducing society" (Fairclough 1992:65), Foucault (1972) asserts that they also have the potential for transforming society, in creative ways.

In many discourses, e.g. interviews, newspaper journalism, the genre is seen as constitutive of and governing the discursive practice. However, I suggest in the case of CMC, as analysed in this corpus, this is not the case, and due to the presence of a number of genres within the corpus, genre, in this instance, does not determine the form of the discursive practice.

Genres

The extent to which oral and literate forms are interwoven in any one CMC utterance is dependant upon the genre of the communication. Yates (1994:180)

The concept of genre used is that theorised by Bakhtin (1986). Everyday language use is made of countless individual utterances (units of speech communication), but there are a smaller number of relatively stable sets of conventions, associated with different types of communicative activity, e.g. interview, chat, or dissertation, called genres. Each genre is, in the context of traditional text or speech-based communication, associated with particular text types and processes of production, distribution and consumption, and with particular modes of intertextuality (Fairclough 1992).

CMC calls for a new application of the concept of genre, theorisation of which is outside the realm of this study. It is evident, however, that trying to view all types of CMC as one communicative genre is misleading; as Yates (1994) explicates strongly, such a view would seriously contradict the idea of the social construction of communication. To say that e-mail, computer conferencing and list discussions, as different forms of CMC, represent different genres, is too simplistic, as each may contain messages with similarities to genres identified in other modes of communication.

In order to identify genres, the corpus was examined and individual messages classified or categorised according to their style of presentation and their content. While one can identify, within this corpus, a larger number of different genres, and some messages could fall within several genres, the messages can be divided into three main genres. These

are discussion of issues, requests for and provision of information, and the formal presentation of information. Table 6.2 indicates the numbers of messages in the corpus falling within these three genres.

Formal information presentation is the easiest of the genres to address and illustrate, and the least common. Messages included a call for conference abstracts and presentation of a cultural assessment tool. These are essentially literate, as opposed to oral, forms, reflecting the formality of literate discourse by closely paralleling their text-based counterparts, e.g. figure 6.5.

Table 6.2. Message genres within the corpus.

a] contribution to discussion

[i] without copying in of text -------- 23

[ii] with copying in of text ----------- 21

    answer to information request ------ 4

c] information presentation -------- 7
(   without prior request)

Requests for information tend to be phrased, and answered, less formally, in a more oral style, for example the request in figure 6.6. Several replies were forthcoming, some providing the information requested, and others reiterating the request, or addressing related issues.

Subscribers to Nursenet seem to be very willing to provide information if they possess or can obtain it, and often at inconvenience to themselves. This phenomenon is not peculiar to nurses (Kaye 1991) and evidence from my own use of other lists indicates that it is currently widespread among Internet users. My own experience of requesting information through the list was that some people replied via the list, others e-mailed me personally, and several others re-posted my request to other parts of the Internet, to try and obtain information that they personally did not have.

Figure 6.5. Example from corpus of information presentation genre.

[header deleted]

RUTGERS
THIRTEENTH ANNUAL NATIONAL NURSING COMPUTER CONFERENCE
THEME: NURSING INFORMATICS ON THE INFORMATION SUPERHIGHWAY

DATES: April 19-22, 1995

Place: Hyatt Regency Princeton Princeton, New Jersey

CALL FOR ABSTRACTS!!!!!!!!

Deadline extended until Monday OCTOBER 24, 1994

FAX ABSTRACT TO:

Gayle Pearson at 201-648-5895

CONTENT:

Focus on use of Internet

Include - Title, Purpose, Objectives, Summary, Implications for descroption of software if a demonstration is to be given.

Contact:

Figure 6.6. Example from corpus of information request genre.

Hi Suzanne....I too graduated from McTavish and live in Buffalo. To those aos of you on the Net, anyone heard of anything to do with chaos theory and nursing. Someone in Ireland a few weeks ago talked about a new journal called Chaos and Nursing. Where can we get access to this? Please answer at IN%[name@address] Thankyou o

The main genre present is discussion, often in more oral than literate form. Issues related to the discourse strategies used are discussed in the text analysis section.

Intertextual chains

Each utterance is filled with echoes and reverberations of other utterances to which it is related by the communality of the sphere of speech communication. Bakhtin (1986:91)

Intertextuality is the degree to which one text is constructed from parts of others or refers to others, although the degree to which this is acknowledged varies substantially. Manifest intertextuality refers to some specific indication of the texts used to form the text under consideration.

Examination of the corpus revealed considerable intertextuality, with many messages containing some or all the content of previous messages cut-and-pasted into them. This feature is not unique to nursing lists; it occurs in many others, as an example from the Derrida list illustrates (figure 6.7), while an example from the corpus illustrates several intertextual strategies (figure 6.8).

This illustrates not only the pasting-in of the previous text to which the author of the 'forwarded message' refers, but the whole of the forwarded message is pasted into the ultimate author's message to the list.

Figure 6.7. Example from the Derrida list of manifest intertextuality using cut-and-paste.

(by convention, >indicates text cut from elsewhere):

Jane M. M. quotes JD:

> 'The "rationality" [...]
> no longer issues from a logos.

Ah, that's the confusion: *a* logos.

> Further, it inaugurates the destruction,
> not the demolition but the de-sedimentation,
> the de-construction, of all the significations
> that have their source in that of the logos.

Note: *the* logos.

> Within this logos [Pre-Socratic to Post-Hegelian]
> the original and essential link to the *phone*
> has never been broken.

Note: *this* logos.

> I only think when someone disagrees with me
> and I have to have another look at my point of view

Oh, that's you writing now
[rest deleted]

This explicit marking of the topic is the least ambiguous method of clarifying a message's context within such multi-stranded, parallel discussions (McCarthy et al 1992). It is a form of intertextuality that is without parallel outside CMC. Frequently the whole of a message to which the author refers is included, which could be argued to be unnecessary, although the most likely explanation is unfamiliarity with the technology on the part of the sender.

Figure 6.8. Example from corpus of manifest intertextuality.

Subject: Re: certification by competency (fwd)
I hope ya'll don't mind me slippin' this in...it seems to be focusing on the areas we are talking about.

---------- Forwarded message ----------
Date: Wed, 13 Dec 1994 16:23:49 EDT
From: Michelle West <west@seattle.ssu.edu>
To: Multiple recipients of list NURSERES <NURSERES@KENTVM.KENT.EDU>
Subject: Re: certification by competency

----------------------------Original message----------------------------
I couln't agree with more Jane. We are unprepared to meet the challenges of the "business" world....but we are learning....

On Mon, 23 Nov 1994, Jane Swift wrote:

> Isn't it wonderful? Certify idiots to perform tasks! Why not a robot?
> After all, we all know that only physicians can evaluate information and
> make decisions. This is one reason we have radiation poisoning in patients,
> the technologists have learned not to see what they see - and go on
> giving treatments after a patient is obviously in trouble. I'm sure there
> are many other situations where information is ignored because we all
> know that only treatments are important....... After all, that is
> what pays the bills.......... Maybe we better figure out in $ and pennies,
> how nurses can decrease medical bills (notice I did not say health care
> expenses...) Forgive my cynicism, but we need to decide what we do and
> why it is cost effective, or more proposals like this silly one will
> become reality. Unfortunately, as nurses we are unprepared in economics.
> I still remember being taught in nursing school (a BSN program too)
> "Nurses do not think about money."
>
> Jane Swift
>

Discourse practice conditions

While the readers (consumers) of CMC messages have only the text itself to interpret (irrespective of however many emoticons the text may contain), the style of the texts varies in a continuum from speech to writing (Yates 1994). A number of factors may influence the style used by individual subscribers, but the primary influence is likely to be technological, tempered perhaps by economic considerations.

Some subscribers may use OLR's (off-line readers) to compose messages/replies. These messages would be expected to be more literate in style, the sender having time to consider the composition of their reply before sending it. Others may have to reply "on-line", and if they are personally responsible for the costs incurred, this would be an incentive to minimise them, so messages would be expected to be shorter, and more oral in style. Others may not have to personally bear the costs (e.g. students/academic staff using facilities such as JANET network), so may have more time to consider their reply and make it more literate.

These questions cannot be considered further here, but could be a useful area for further study.
 

Text analysis

While an exhaustive text analysis is not feasible, examination of the corpus revealed a number of features that seem pertinent to nursing discourse. These are addressed in the following sections, and include politeness and repair, modality, and interactional control, including the issue of turn-taking.

Politeness and repair

Politeness refers to a set of practices used by discourse participants to mitigate 'loss of face', or to promote their 'positive face', e.g. to be liked or understood. Fairclough (1992:163) views investigation of politeness strategies as a useful way of "gaining insight into social relations within the practices and institutional domains" with which they are associated.

The potential within CMC for misunderstandings has been discussed earlier, and within this corpus, nurses seem to use a high proportion of politeness mechanisms. This may be an attempt to defuse the potential for misunderstandings they recognise exists, or may be a feature transferred to this medium from nurses' everyday speech-based communications.

Examples from the corpus include:

1] I really think you have some good points here, but..

2] I beg to differ and wonder why...

Closely allied with politeness is the concept of repair, i.e. attempts to correct or forestall misunderstanding within communications. Repair seems to occur less frequently than one would expect necessary if the misunderstandings for which many researchers have identified potential actually occurs. This may in part be explained by the strategies used by some contributors, which can be termed prophylactic repair. For example, one contribution to the corpus ends with the sentence:

Really John, I'm not trying to stir anything up. I don't know you and I won't attempt to generalize to all educators as I know everyone is unique. But I just speak to what I have personally seen here.

Repair is not peculiar to CMC, and one can see similar mechanisms employed in FTF discussions, as well as the more familiar form of repair, where clarification is used:

I was not stating that nurses who receive higher degrees are elitist, but rather that to state that only they know the field and should dictate the policies and procedures are elitist.

Figure 6.9 Examples of emoticons or 'smileys'.

View by turning paper clockwise through 90 degrees

:-) Smile
:-( Frown
;-) Smile with a wink
8-) Smile, wearer has spectacles

Modality is the degree of affinity with or commitment to a proposition expressed by an author (Fairclough 1992). In grammatical terms it is associated with modal auxiliary verbs such as 'may', 'should', 'must', and, more importantly, with adverbs such as 'possibly', 'obviously', and with hedging of assertions, e.g. 'sort of'. Fairclough (1992) sees modality as being associated with power relations, extensive use of modality (indicating low affinity with statements) being associated with lack of power. Modality is much more central to power relations in discourse than has been acknowledged, and would be an ideal area for further study.

The use of modal phrases such as "er" or "sort of", is rare in this corpus, as opposed to the frequency with which it is seen in spoken discourse (figure 6.10).

Figure 6.10 Extract from interview transcript with A&E nurse.

Yeah. I think it, sort of, provided very holistic care for the elderly lady coming through the unit, who actually gained more benefits than simply having a wound dressed on 'er leg. Erm, I think that had it, had she 'a' been seen in an ordinary unit without nurse practitioner cover, the chances are that the, er, medical staff there would've dealt with 'er leg. .. And, you know, that's not being derogatory to the medical staff, but I think that their purpose is there to deal with the injury, etcetera. An' I think that perhaps the nurse, because of the holistic nature of their work, can actually broaden out a little bit more. Because they have that focus only on the patients with minor injuries etcetra. They haven't got other priorities within the unit at that time. (Murray 1994)

Nursing is the only field in which massive numbers of workers within it ...

I really think you have some good points here, but my feeling is...

For what it's worth,...

Maybe because nursing is ...

I think an answer might be ...

Analysis of interactional control, as defined by Fairclough (1992), considers how and by whom interactions within a discourse are controlled, whether control is negotiated or imposed, and asymmetry of interaction between participants. It includes consideration of turn-taking mechanisms (Sacks, Schegloff & Jefferson 1974), and the introduction and development of discussional topics. Turn taking and its existence in CMC have already been discussed (Chapter 4). Sacks et al (1974) and Fairclough (1992) suggest that all participants have equal opportunity, although Fairclough also recognises that in practice, this does not occur, for reasons including context, genre, gender, and power relations.

Analysis of this corpus indicates that a form of turn-taking occurs, but peculiarities of the medium restrict which of the rules can apply, the usual form being self-selection by participants. One major factor is the simultaneous existence of multiple discussion threads, any of which might be addressed in any turn by a next participant. Technical factors also mean that, even if a list message is directed at an intended 'next participant', others may read it and respond first, which may also change the direction of the discussion. A similar problem exists for subscribers who receive messages in daily digest form, as by the time they have received and read messages, the discussion may have moved on by several turns. This does not prevent return to an earlier point in the discussion, usually achieved by making explicit reference to the earlier point using manifest intertextuality.

Messages are usually opened in such a way as to address the whole list, although some have openings directed at named individuals, indicating that the message is primarily a response to an earlier message from a particular participant.

It is rare for participants to continue in the absence of a next participant within a topic thread, i.e. if a message receives no responses, it is unusual for the sender to pursue the point.

Overlap, i.e. two or more participants contributing simultaneously to a conversation, is observed in FTF or telephone conversation, but is rare. It can be used as a power mechanism, but Levinson (1992) estimates it occurs in less than 5% of the speech stream. Technical aspects of the medium mean that overlap is impossible in CMC.

One feature which renders CMC akin to written text and some formal speech (e.g. lectures) is that the size of turns is governed entirely by the participant. In FTF communication opportunities may exist for a next participant to interrupt and govern size, but unless technical failure truncates a contribution, participants have total control over the size of their contribution. Cost and other factors, discussed elsewhere, may limit this in practical terms.

Having discussed turn taking, it is useful to consider other aspects of topic control. As the list is unmoderated, any individual can, in theory, make any contribution, or introduce a new topic, at any time. Whether the topic is taken up by other subscribers varies, as does further contribution by the person introducing the topic. It varies with the nature of the topic, so a request for information may produce replies, focused on the original request, or may precipitate a wide-ranging discussion around the topic or a tangential issue. Some topics are short-lived, with few messages, while others range over a considerable number of turns. This is an important difference from FTF discussions, and relates to some of the criticisms made by Walther (1992) of experimental studies of CMC.

In the corpus, three new topics were introduced, of which two are briefly discussed to illustrate the types of responses possible. The first, from a nurse in Europe, concerns the use of initials in messages and presuppositions of subscribers understanding their meaning (figure 6.11). The author has used manifest intertextuality by cut-and-pasting part of the message to indicate context. There were 3 responses in the corpus, one from Europe and two from North America, one simply listing the initials in question and what they meant. This stimulated a further response two days later from a European nurse who, while thanking the provider of the terms, pointed out that the reply didn't provide any information about what the terms entailed. This topic then had no further messages.

Figure 6.11 Example of topic introduction.

> IMHO the ANA is only interested in the ANA. Their main push (that I'm
> aware of) is that NO one is to be considered a *real* nurse unless they
> at least have a BSN. What a waste! When we're talking about nursing
> we're talking about PCT/CNAs, LPN/LVN and RNs.

The Internet is not limited to North America. Organization and
education/titles differ, several of the initials above don't mean
much to me.

Figure 6.12 Example of request for information.

>To those aos of you on the Net, anyone heard of anything to do
>with chaos theory and nursing. Someone in Ireland a few weeks
>ago talked about a new journal called Chaos and Nursing.
>Where can we get access to this? Please answer at [e-mail address]

The second new topic, which was a request for information, illustrates the use of several strategies. After opening with a personal greeting to another subscriber who had posted a message previously, it continued with the sender locating herself in the message and the community of the list, before presenting the request (figure 6.12). Several responses were forthcoming, the only one in the corpus being from another subscriber indicating they would also like the information. Over the following two days, four responses were posted, one of which provided the precise information requested, and two providing some relevant information.

Within this list, it seems that, when information is requested, once it has been provided by one or more people to the list, the topic ceases to be of interest to the list, unless discussion is promoted around the issue. Similar levels of response were seen with other information requests outside this corpus. What cannot be ascertained from the corpus is how many people replied to the senders of messages using personal e-mail. In my experience of posting a request for information, I received replies via the list and several personal e-mail messages with the information. One explanation might be that providers of the information feel that it may be of interest to others on the list; this issue cannot be resolved at present, but may merit further study.
 

Social practice

The structure of a language is a social product of our language faculty. de Saussure (1915/1988)

Fairclough (1992) concentrates on issues of ideology and hegemony, drawing on the work of Althusser and Gramsci, in discussing discourses' social practices. If ideology refers to the dominant ideas supporting a social practice, then DA seeks to identify the role of discourse in supporting or opposing the ideology. As Fairclough (1992:90) points out, one cannot assume that "people are aware of the ideological dimensions of their own practice." Gramsci's concept of hegemony deals with the exercise of power by one group within a social setting. While Foucault (1980) finds ideology problematic, and so its inclusion in the model could be seen as inconsistent with the overall philosophical basis of Fairclough's method, the model is still useful for examining power relations and whether particular discourse practices "reproduce, restructure or challenge existing hegemonies" (Fairclough (1992:95).

I have chosen to examine briefly the role of CMC in creating communities and on-line selves, some aspects of power relations and the relation of CMC discourse to supporting or challenging general nursing discourse.

Community and self

Who are we when we are on-line? Jones (1995:15)

The literature on social aspects of CMC has been discussed (Chapter 4), including criticisms of the generalisability of findings from work suggesting reduced socialisation (Walther 1992). The social construction of reality via computer networks is discussed by Jones (1995), who believes that CMC allows for customisation of social contacts from fragmented communities, and that it fosters community among users. Baym (1995:141) has also examined CMC groups where "participants have voluntarily created communities rich in social information, prominent personalities, valued relationships, and behavioral norms". She asserts that styles of communication or discourse are related to social practices existing outside CMC, which are then imported into the medium. Jones (1995) believes that the connections between the on-line community and the wider social practice of which participants are a part is currently unclear.

Evidence relating to all of these issues is apparent in the corpus. Behavioural norms are one important feature of a community, and while there is little evidence of overt enforcement of norms, participants in the list seem to conform to aspects of "netiquette" such as politeness, and using the list for the purpose for which it is meant, i.e. as a "global forum for nursing issues".

The wider social practice of nursing, within which most contributors are encultured before using CMC probably influences communication styles, as has been discussed earlier. One currently unresolved issue is whether the dominance American and Canadian nursing issues within the list's discussions risks becoming translated into a North American hegemony, with erosion of diversity in practice and theory world-wide.

Many nurses who post messages to the list reveal a great deal about themselves, sometimes when introducing themselves to the list (a convention), but often using their own experiences as examples to illustrate issues under discussion, although it is not possible here to resolve the issue of whether this is a feature of the list, or of CMC, or whether these are people who would make this kind of self-disclosure in other situations.

Within Fairclough's method, it is important to consider whether the discourse supports or challenges social practices. It appears that, generally, CMC discourse challenges prevailing ideas within nursing, although this is not to suggest that this does not also occur in other nursing discourses. Many subscribers wish to change nursing from its present state, and many postings discuss ways in which practice or education might be improved, or new theory generated. Many nurses post messages asking for ideas or information about ways in which they might improve their own practice, and this thread also appears strongly in the responses of the interviewees. This is appropriately summarised by one interview participant, who said:

I think that nurses who use the Internet are generally the forward thinking ones. But, usually the ones without much power. The Internet provides a voice for many of us who do not get listened to at our own institutions. It gives strength to find out that one is not the only one who has such "outlandish," ideas.

An important area for future research will be the influence creation of an international community of CMC-literate nurses will have on other aspects of nursing practice, education, theory and administration.

Power/ideology

CMC minimises through the anonymity of users the functioning of power relations. Yates (1994:136)

The potential for CMC to subvert or oppose traditional power structures and allow the "powerless" greater access to communication media has been discussed from the literature. Power relationships, from a postmodern/poststructuralist perspective such as Foucault's are viewed as not natural or objective, but as "socially constructed intersubjective realities" (van Dijk 1985), maintained through discourse.

Discussion within this corpus seems to support the contention that CMC has equalising potential, and is capable of providing greater opportunities for the traditionally powerless, in this case, women. Over a 5-day period, including the two days containing the corpus, there were 28 contributions to the discussion thread from 10 males and 33 contributions from 18 females. This is a greater contribution from females than might be the case in a FTF discussion with similar proportions of contributors. The original contributor (male) of the message starting the discussion only made 3 contributions (of 61) to the discussion, less than might be expected in FTF discussion, where he might feel more compelled to explain and defend his position.

The proportion of contributions from males (46%), even taking into account that 10 of the 28 were from one person, is much higher than the proportion who are subscribers to the list (approx. 21% (Norris 1995)). As males are more willing to 'voice' their opinions in this forum than females, this could be viewed as domination of the discussion, as is frequently seen and criticised in FTF discussions. One factor that might have influenced the proportion of males and females contributing was that the original message was posted by a male. A comparison with a similar discussion initiated by a female might elicit whether similar patterns of contribution are seen. The high proportion of contributions from males may reflect their traditionally greater familiarity with computer systems and with the conventions of netiquette (one topic per message). In considering any wider applicability of these findings, one must consider that, while 80% of Nursenet subscribers are female, less than 5% of Internet users are female (Grant 1994).

It has been suggested that communication is equalised because of the absence of the visible trappings of power and influence. One aspect of messages that has received little analysis, but which merits a brief discussion here, is the use of "sig files", i.e. signature files. Baym (1995:156) suggests that they are "are one of the most immediate and visually forceful cues to identity". In considering the two examples below (figure 6.13) of types of signature files that might be used, one can see the potential for the author of a message to attempt to exert authority within a discussion.

Figure 6.13 Two examples of 'sig' files, or signature files.

Sig file 1:

Peter Murray
petermurray@delphi.com & pmurray@liv.ac.uk

Sig file 2:

*************************************************************
*Peter Murray, BA(Hons), RGN, RNT
*Nurse Tutor, Chester College School of Nursing & Midwifery,
*Chester, UK
*Tel:0244 383688
*E-mail: petermurray@delphi.com & pmurray@liv.ac.uk
*
*"il n'y a pas de hors-texte" -- Jacques Derrida
*
************************************************************

One interpretation, and others undoubtedly exist, might be that by using sig file 1, I am merely identifying myself as a contributor to the discussion, albeit also indicating gender, and providing anyone who wishes to contact me directly with an address to use. I am not even indicating that I am a nurse, and have a valid reason for contributing to discussion on a nursing list, although the e-mail address indicates country and that Internet access is from an academic institution, but not whether I am a student or teacher.

Sig file 2 could be interpreted totally differently; by including my job title, qualifications, and place of work, one could argue that I am saying "I have this authority to speak". The use of such an elaborate sig file, including a quotation (a common feature of sig files), also suggests that I am sufficiently computer-literate to know how to produce one and append it to messages.

Sig files were not used frequently within this corpus. Most contributors simply give their name and e-mail address (a netiquette convention, as not all systems distribute this information in message headers). This may indicate lack of computer literacy, or it may be that must subscribers do not see value or utility in sig files, or that they do not wish to try and exert any authority to speak within their message, but to contribute as an equal. Sherblom (1988), analysing sig file use in e-mail within an organisation, suggests that they are used more frequently by subordinates sending mail up the organisational ladder. Within the corpus, sig files tend to be used more frequently by males.
 

Multiple threads to discussion

Figure 6.14 Examples of message headers used in the corpus.

Re: ANA
Elitist comments or..the Language of Nursing
Re:Elitist comments or..the Language of Nursing
An honest question
Re:An honest question

An additional subject header, had a digest subscriber pressed 'reply' would have been:

Re:NURSENET Digest - 13 Dec 1994 to 14 Dec 1994

Contributors use essentially three mechanisms for indicating to which discussion thread they are contributing. Firstly, there is the 'Subject' section of the message header, although this is not always reliable, as senders may not give a subject, may provide a default header by using the 'reply' button on their mailer, or may type in a new header. Examples from the corpus indicate the following variety of headers for just one discussion thread (figure 6.14).

A frequent method within FTF conversations for indicating return to a contribution prior to the previous turn is to say, for example:

As John said a few minutes ago....

I agree with Sue's comments, but...

This device is the second used in CMC, although tends to be used infrequently, due to the ease with which the third method can be used (or sometimes misused) by most subscribers. This, most frequently used, method is the pasting-in of all or sections of previous messages, i.e. manifest intertextuality, which has been discussed in Chapter 6.

Chapter 5

Chapter 7

Contents

Copyright: Peter J. Murray 1996, 1997, 2000

Click here to return to différance, Peter Murray's Home page

Created and maintained by:  Peter Murray
E-mail: peter@nursing-informatics.net
Last updated 23 June 2000 (contact email address changed; previous address on longer valid; previous job affiliation deleted)
Previous update: 21 June 1997 (design only changed; substantive text unaltered)