October 1996

Feeling Between the Lines

by Joyce Menges

Considering that effective communication is paramount to success both in personal and professional endeavors, and considering the exponential growth of the metaphoric lines of communication in cyberspace, the author has embarked on this path of inquiry: What do we need to know about nonverbal communication in order to succeed in our interactions within the emerging connections made possible by the growth of electronic communication? Some would argue that human intercourse, often a significant challenge even in face-to-face human interaction, is severely restricted by technology. While effective communication has many aspects, extraverbal cues seem to be the most elusive. Using a combination of quantitative and qualitative research techniques, the author has described her personal experiences and survey research into synchronous IRC and MOO communication.

Undoubtably, there is communication in the sychronous environments of cyberspace that goes beyond verbal conversation. Gaining an understanding of, and a greater capacity for expressing and perceiving extraverbal communication is practical for all who use computer-mediated communication to play and to work.


Communication is the fundamental requirement for success in personal and professional endeavors of all kinds, at all levels (Chadwick & Lind, 1996, p. 57). Arguably, this simple statement is true; if we cannot know each other we cannot succeed. Chadwick and Lind were addressing their remarks to high school students preparing for admission to technical programs of study and beyond to entry into the work force, but their words surely hold true for the rest of us. Given this, what do we need to know in order to succeed in our interactions within the emerging connections made possible by the growth of electronic communication? While effective communication is often a significant challenge in face-to-face human interaction, some would argue that it is severely restricted, perhaps even crippled, by technology. My purpose in undertaking the work following has been to discover and describe one vital and seemingly elusive (non-existent?) aspect of the process of human intercourse in the textual environments of electronic communication, namely the implied messages of extraverbal communication in cyberspace.

Consider the concept of lines of communication. The military once struggled to keep lines of communication open with semaphores. We speak metaphorically of reading the messages between the lines. And, of course, innumerable messages travel phone lines daily. In the text-based world of cyberspace, where human communication transpires without benefit of the physical presence of the participants, are there lines of communication that transcend the written words exchanged? Are there messages lurking between the lines of text? Does communication in the synchronous virtual spaces of Multi-User Object Oriented (MOO) environments and the other forms of Internet chatting permit extraverbal pointers to emotions and attitudes? If so, what are the most significant ways humans express and perceive feelings in cyberspace? What follows is a narrative outlining my first attempt to answer these questions. (1)

A Preliminary Definition

A survey of the communication research literature has revealed that during the late 1960's and early-to-mid 1970's there was an explosion of research and writing on the topic of nonverbal communication in face-to-face human contexts. It is not merely a hidden dimension or a silent language that has been uncovered by a new wave of scientific explorers; wrote Montagu and Matson (1979), it is more like a neglected universe of discourse and intercourse. We are becoming aware that the verbal domain is only the part of communicative expression--there is much much more to the human dialog than meets the ear. (p. xiii) Now, from Erving Goffman's early work in The Presentation of Self in Everyday Life (1959) to Nathan Joseph's more recent work Uniforms and Nonuniforms: Communication Through Clothing (1986), seemingly every conceivable angle from which one might view the topic has been explored. That is, of course, every angle but the present one.

For every primer written on the topic there has been a corresponding definition, theory and approach to the study of this phenomenon. In fact, the term itself has been the subject of considerable debate and reforming. We prefer the designation implicit communication to the commonly used misnomer, nonverbal communication, states Mehrabian (1981) as he grappled with his sense of dissatisfaction with the popular cultural term for something he described as the transmission of information about feelings and likes-dislikes or attitudes. (p. 3) Neither does nonverbal accurately describe these same things as they occur in cyberspace.

Certainly, since one definition characterizes verbal as being "of or in words" (Ehrlich, Flexner, Carruth & Hawkins, 1980, p. 770), electronic communication, which is almost entirely text-based, is verbal. Here too, nonverbal is a misnomer. For the purposes of this work, I use the term extraverbal communication to refer to the non-dialog devices employed in human communication in sychronous computer-mediated environments to reflect thoughts, feelings, and reactions.

Building a Theory

One early nonverbal communications researcher, Mark Knapp (1972), has made the following estimate: considering that the standard spoken sentence takes only about 2.5 seconds, the average person actually speaks words for a total of only 10 to 11 minutes daily. Given this, the verbal dialog portion of an interaction will carry less than 35% of the social meaning of any particular situation. More than 65% will be carried nonverbally (p. 12). Other researchers would, however, disagree. “In part, [intellectual confusion] stems from a lack of agreement on the boundary between verbal and nonverbal and the distinction between communicative and non- communicative behavior” (Harper 1978, p. 3). Still there is tacit agreement here; regardless of the terminology or the parameters, there is an extraverbal component to human communication. It is my argument that this also holds true for electronic communication.

Some have narrowed their investigations to one discrete class of extraverbal cues. Nathan's (1986) work related to clothing for example, most have organized their study of nonverbals around one of two approaches, behavioral categorization or functional analysis. Categories of nonverbal behavior surveyed have included some combination of the following: touching, proximity, orientation, appearance, body movement (posture, head nods and gestures), eye behavior, clothing, smell, taste, paralanguage, nongrammatical aspects of speech such as tone, accent, speed volume, artifacts, and environment (Barker , 1990; Druckman, 1982; Hinde, 1972; Knapp, 1972). Functional theorists have, on the other hand, identified purposes for nonverbal communication which range from simple schemes like Hinde’s (1972) three functions: managing the immediate social situation, sustaining verbal communication or replacing verbal communication, to more complex conceptions such as Druckman's (1982) five functions which include such purposes as "Indicates pre-articulate feelings, provides clues to information processing, serving as emphasis in persuasive appeals, facilitates deception or holding back of information, and conveying subtle messages." (p. 19) A more common list, first articulated by Knapp (1972; see also Barker, 1990; Patterson, 1983) includes repeating, contradicting, substituting, complementing, accenting, relating and regulating. Mirroring the work of the face-to-face theorists, I have sought to discover and describe extraverbal devices and their function in text-based electronic communication.


Unlike applied research which seeks to offer explanations for a broad population using quantitative analysis, the action research methodology used here is focused on a narrower context providing useful clues to understanding a particular arena. This understanding may or may not represent the broader setting. Action research is frequently a framework for description; combining several methods of data gathering, both quantitative and qualitative; this methodology provides checks and balances to an otherwise nearly nominal form of research by a process referred to as triangulation. As well, action research is by nature, holistic; the researcher paints broad strokes about the topic under consideration, gaining perspective on the whole rather than its particulars. For this effort, I have combined the quantifiable data gathering process called Nominal Group Technique with a purely qualitative description of myself as participant-observer in order to achieve a description of extraverbal communication in two of the sychronous venues of cyberspace communication.

Researcher's Notes

My observations as researcher have grown out of my interest and participation in computer mediated communication these last five years. As my experience has grown I have become increasingly aware of my ability to share not only my words but my feelings with others and perceive the same in return. My commentary here stems entirely from this experience both before I began this formal study and, more pointedly, since. The nominal groups, on the other hand, were my effort to gleen the best thoughts of other experienced virtual communicators but have also provided data for more careful, studied observations through the logs of each session. (2)

Nominal Group Technique

The process called Nominal Group Technique (NGT), developed by Andre L Dalbecq and Andrew H. van DeVen in 1968, is a research tool used to facilitate organized group brainstorming about a single research question. While it has elements in common with a survey instrument and a focus group, it is neither of these things. Through a rather regimented but effective series of steps (See Appendix B) the facilitator leads the participants towards the final goal, to develop a set of ideas that the group thinks is important to the research question under consideration.

The NGT participants gathered for this study were a convenience sample of twenty-seven of the author's colleagues, acquaintances and friends from two cyberspace venues, a Multi-user dimension, Object Oriented (MOO) virtual space and Internet Relay Chat (IRC) gathered into four Nominal Groups, two in each arena. The research question posed to the participants was: "What are the most significant ways humans express and perceive feelings in the synchronous environments of cyberspace?"


As I have said, this work has arisen out of my personal experiences in cyberspace and my own need to better understand what I knew intuitively; there is more to human communication in cyberspace than the mere words of conversation. As I have worked and played in this virtual place I have witnessed learning and intellectual discourse, commerce of every imaginable ilk, the social interaction of strangers and friends, enemies and lovers, and chatter, idle and otherwise. In that time I have come to know that human communication across the lines of the Internet can be rich with extraverbal clues to emotion and attitude.

Consider the following exchange:

[Joyce] I have to go shortly to prepare for that.
[Joyce] I have had my nose in my computer all morning
[Joe] i can see the smudge on the screen, yes indeedie.
[Joyce] hehe, no way!
[Joyce] here, let me fix it...
[Joyce] *wipe wipe*
[Joe] yes you did... well... i would love to keep you here for ever, but i think if i had my choice.. i would feel better knowing you had sufficient time Joyce to prepare, even if that meant not spending a few more minutes with you.
[Joe] all that to say...
[Joyce] it's ok if I go even tho you want me to stay? :)
[Joyce] *grin*
[Joe] words mean things, as you and I know... and i can't give you a wink or a tone or a facial smile to let you know it's fine to go... so I have to say it in a way I am convinced the meaning gets through the smudge still on my screen, :).
(J. Subjak, personal communication, 2-3-96)

Unintentionally here, my friend Joe has verbalized one of the primary ways people express their feelings in the virtual realm; they talk about them. But the exchange also reveals other devices that are typically found in conversations of this type. Actually, the Nominal Group participants identified at least a dozen distinct ways in which people interact extraverbally. Of course there were many devices mentioned during the group process and probably no one particular participant would subscribe to the following list just as it stands, but I believe it is an accurate reflection of the collective groups' best ideas and it does echo my own experiences. (See Appendix D for a comprehensive summary of NGT results.) The NGT participants identified five significant ways they believe people express and perceive extraverbal communication in cyberspace: emotes, word choice, punctuation, emoticons and capital letters. A discussion of these follows.


The fact that MOO and IRC programmers have gone to the trouble of writing code to facilitate the use of emotes supports my original hypothesis: people need to, want to and do act extraverbally in cyberspace. But wait a minute; this preliminary statement is fundamentally flawed. People don't act in cyberspace, do they? Doesn't the word act imply physical motion? I should be saying people represent actions they would take in face to face contexts, when communicating in these virtual venues, using textual descriptions. To save us all from the verbal gymnastics of that last sentence, however, I ask that you to read what follows with a bit of suspended disbelief. When I use words like "behaves" and "acts" what I really mean to say is represents behaviors with text.

It is easy to find examples in the NGT logs of emotes that serve Knapp's functions. Nods, for example, are used to reinforce (repeating) affirmative responses or in place of them entirely (substituting). In the following brief exchange from the second MOO NGT there is a bit of regulating:

Joyce says, "is a pretty good list :)"
JackG writes this on the whiteboard:
>> commanding statements
Joyce asks, "anything else?"
Joyce looks around the room carefully
Anne says, "Not from me"
JackG says, "I'm finished"
Joyce nods to Anne (personal communication/NGT log, 3-24-96)

The participants had gone fairly silent by this point in the group process but I wanted to make sure everyone had listed all their ideas. "Joyce looks around the room carefully" was my way of making room for any last minute ideas while beginning the transition to the next phase of the group process.

Apparently, however, emotes have not been enough to satisfy the virtual space communicator because other extraverbal devices have emerged that also serve to transmit clues to feelings and attitudes.

Punctuation, Emoticons, & Capital Letters

There is much to be conveyed in the turn of a bit of punctuation or case, I have learned. An exclamation point can transform a simple statement of fact into a declaration, often changing the whole flavor of the sentiment transmitted. Can you feel the difference between someone saying, "I'm swell." and "I'm swell!" In the first instance you would be wondering if the person was being facetious, in the second you are quite sure of his mood. An experienced virtual communicator might interpret the ambiguity of the first statement as an effort to deceive or reveal a mood or attitude not articulated. The second simply reinforces the obvious words. As well, upper case letters serve to accentuate certain words or phrases in an exchange. You cannot raise your voice in a virtual conversation but you can RAISE YOUR VOICE! And your favorite smiley emoticon can entirely transform the implied meaning of the words used. Sometimes, just as in face-to-face communication, it is difficult to tell when someone is joking. In the conventional venue, you can catch a twinkle in the eye that reveals the teasing nature of the words, "You're awful." In the virtual world you might be offended by these same words were it not for the emoticon that punctuates them ("You're awful! :-)")

The use of ellipses is particularly interesting. Since all such conversations are typed, there is a natural lag to the flow of conversation as thoughts are transformed into text. While it is common to trip over each others words from thought to thought in a way that becomes quite natural and acceptable, strategically placed ellipses dots help to regulate virtual conversations. Punctuating the end of a line of text with these three dots reminds the reader that there is more to the thought coming on the next line. They can also simulate the natural pauses in audible speech as in the following exchange from the first IRC NGT:

[Joyce_] what about that fellas... is Brent correct?
Joyce_ waits while you talk among yourselves :)
[AllenR] I think so... yes
[Da_Bear] sure
[rrider] Yes, I agree with that

(personal communication/NGT log, 4-6-96)

Word Choice

Ultimately, however, the most powerful transmitter of extraverbal information is words. As great literature has shown us repeatedly, carefully chosen words can convey the most powerful of emotions with perfect clarity. A particularly lively discussion developed around this very idea in the last of the NGTs. Read (3) this excerpt and watch for the use of well chosen words as transmitters, and the content of the exchange, for this group's thoughtful debate about this issue:

[Joyce_] everyone agree the list is good enough to vote on?
[lizzie] no
[Joyce_] please feel free to discuss further guys :)
* Joyce_ wait and watches with interest
[JmpMstr] voting to find the top six???
[dori2] lizzie describing actions is more than just making actions - it is a more detailed thing
[Joyce_] JM: yes, that's the next step
* OldBear wishes there were a chalk board or place we could write each item on yellow stickies and then cluster them into groups. This medium is not very good for developing true group hubris. ;)
[lizzie] dori, convince me
* JmpMstr lines up with OB... I agree
[dori2] well I can say dori2 sits
[lizzie] dori.... you mean... dori sits on the chair hapazardly
[dori2] that is simply using /me command
[dori2] yes - that is closer lizzie
[lizzie] ooooooooooooooooo, this group is creating a mind meld... we both knew you were sitting
* lizzie gets spooked
* Joyce_ smiles
[dori2] it is more detailed - that gets the behavior more clear
[Joyce_] what a great group!
[lizzie] wait.. i have to think it out
[lizzie] Will... waht about you.. i picked on two of yours
[lizzie] do you have convincing arguments????
[Joyce_] wanna see the list again?
[dori2] using /me command is part of the expression - detail is describing more
[lizzie] you SAID THAT
[JmpMstr] remember that we're talking about expressions AND perceiving... what is it that helps you perceive????
[dori2] the more we type, the better chance we have at perceiving what the other is trying to say
[lizzie] but how do you SHOW your perception????
[lizzie] ooooooooooooooooooo, William.... WELL done!
* OldBear perceives lizzie's point.

(personal communication/NGT log, 4-10-96)

When OldBear expresses his wish for a chalkboard upon which to organize his thoughts, for example, it is easy to sense his frustration with the limitations of the medium. It is also easy to believe that this person is taking his task as participant in the NGT process seriously and thoughtfully. He did not have to express exactly these things, but they are available for those who can hear them.


It is not possible to make grand generalizations about extraverbal communication in cyberspace on the basis of this present work. The data is too specifically tied to me and my experiences. Even the demographic characteristics of the NGT participants are skewed in favor of the people with whom I live and work, virtually, and cannot possibly be held up and a truly representative sample of the population. The nominal groups have confirmed, however, that my experiences have been commonly shared among the participants. I submit that this makes the present work an adequate foundation for further study and I offer it as such.

There are other significant difficulties not yet addressed. In particular, I think veracity is a problem. How can one know with certainty what another person is really like or what he is really thinking in these virtual exchanges? The simple answer is: you cannot. Even in face-to-face communication there are no absolutes about nonverbal clues and no one can interpret even the most obvious of clues perfectly. However, time and experience are great teachers in both venues and can, I believe, lead to very accurate interpretations.

All of this virtual communication has relied upon the speaker's facility with words. An inability to type, a language or cultural barrier, or low skills in expressing oneself textually are hurdles to be overcome in computer mediated communication and not everyone is up to the task. Undoubtly anyone working from these deficits will be misunderstood on occasion. Practiced virtual communicators will, however, recognize this and adapt.

For this work, I have looked at only the most obvious devices. What of the more subtle devices employed in virtual communication of personality and emotions? In particular, I am fascinated by the way people dress themselves virtually. Just as a uniform can tell you much about the wearer, an IRC nickname can, seemingly, reveal a great deal about the person using it. Examining this one subtle message-sender could easily hold my attention for a significant period of study.


There is communication in the sychronous environments of cyberspace that goes beyond the mere words of conversation. But why is that significant? On the eve of an age when these exchanges promise to become commonplace in our global culture, gaining an understanding of, and a greater capacity for expressing and perceiving extraverbal communication is practical for all who use computer mediated communication to recreate and to work. Charged with providing support services at a distance via the text-based synchronous environments of the internet, this capacity is of paramount importance. This effort has, hopefully, begun the journey toward that understanding and capacity.


  • Barker, L. L. (1990) Communication (5th edition). Englewood Cliffs, New Jersey: Prentice Hall.

  • Chadwick, M. & Lind, J. (1996). Why Do I Need to Know That? A guide to competencies leading to technical careers in Southern Maine. South Portland: Southern Maine Technical College, Tech Prep Program.

  • Druckman, D., Rozelle, R. M. & Baxter, J. C. (1982). Nonverbal Communication: Survey, Theory, & Research. Beverly Hills, California: Sage Publications.

  • Ehrlich, E., Flexner, S. B., Carruth, G., Hawkins, J. M. (1980). Oxford American Dictionary. New York: Oxford University Press.

  • Harper, R. G., Wiens, A. N. & Matarazzo, J. D. (1978). Nonverbal Communication: The State of the Art. New York: John Wiley & Sons.

  • Hinde, R. A. (Ed.) (1972). Non-Verbal Communication. Cambridge, Massachusetts: University Press.

  • Joseph, N. (1986). Uniforms and Nonuniforms: Communication Through Clothing. New York: Greenwood Press.

  • Knapp, M. L. (1972). Nonverbal Communication in Human Interaction. New York: Holt, Rienhart & Winston, Inc.

  • Mehrabian, A. (1981). Silent Messages: Implicit Communication of Emotions and Attitudes (2nd ed.). Belmont, California, Wadsworth Publication Co.

  • Montagu, A. & Matson, F. (1971). The Human Connection. New York: McGraw-Hill Book Co.

  • Patterson, M. L. (1983) Nonverbal Behavior: A functional Perspective. New York: Springer-Verbig.


  1. By necessity, this work will contain many terms and expression that are new to the reader who is unfamiliar with electronic communication. Please see the glossary for definitions as needed.

  2. Each group session provided a recorded transcript which was used for observation and analysis.

  3. For some readers, these excerpts from the NGT logs might be difficult to follow; the exchanges frequently come in incomplete sentences that bounce off a thought from another participant. Think about the following as you read.

    Often, the speaker will direct a comment to a particular person by first typing their name or an abbreviation that represents their name. Use this to help you sort through the different threads of the conversation as they unfold. Sometimes it is also useful to read several lines from one speaker to get his complete thought, disregarding, momentarily, the interwoven lines from other speakers.

    This filtering process, this sorting through the mixed up lines of text, becomes second nature with time and experience, but can be very daunting to the uninitiated.

Appendix A - Glossary of Terms

Asynchronous In this context, asynchronous describes communication between individuals or groups of individuals using email, bulletin boards, web pages or other static textual internet exchanges.

Character The login name used by individuals to identify themselves on the MOO. On IRC the equivalent would be a persons nickname.

Emote As a verb, emoting refers to using messages that represent action not conversation - 'Joyce smiles,' for example. The term is also used as a noun referring to these actions.

Emoticon A series of ascii text characters used in sychronous communications to express emotion. " :-) " for example is the traditional symbol for a cyber-smile. (Tip your head slightly to the left to see the face.)

Extraverbal Communication While the definition is still a work in progress, I found myself needing a new term to name the thing I am trying to describe in this study. Extraverbal communication refers to non-dialog devices employed in human communication in sychronous computer-mediated environments to reflect thoughts, feelings, and reactions.

Internet Relay Chat (IRC) IRC (original programming code written by Jarkko Oikarinen) is a multi-user, multi-channel chatting network. It allows people all over the internet to talk to one another in real-time.

MOO MOOs (Multi-user Object Oriented virtual environments) are text-based, virtual reality sites that allow people to connect to the same place at the same time and permit synchronous communication.

Synchronous In this context, synchronous describes communication between individuals or groups of individuals that occurs in real-time (as in face-to-face conversation).

Appendix B - The Nominal Group Script


[1] Silent generation of ideas
[2] Round-robin recording of ideas
[3] Discussion for clarification
[4] Preliminary vote on item importance
[5] Discussion of the preliminary vote
[6] Final vote


Please take some time to think quietly about the following question.

"What are the most significant ways humans EXPRESS and PERCEIVE feelings in the synchronous environments of cyberspace?"

Now, write down your ideas regarding this on a piece of paper. Use brief phrases only, please; you do not need complete sentences to capture your best thoughts. For example "using the /me command" is one idea. Let your thoughts flow without constraint or inner debate. There are no wrong answers. And, please, work independently for now. There will be time for discussion a bit later on. I will call "TIME" in about 5 minutes.


Now is the time to get the group's ideas out "on paper." What's important here is that everyone finds voice for their ideas. I will ask each of you to type out your ideas to the channel, one idea per line. Remember: short phrases please. Repeat until you have recorded all of your ideas. Please feel free to add new ideas as they occur to you even if they are not down on your paper. Also, please continue with all your ideas even if you think they may be duplicates of ideas already presented. The group will weed out the redundant thought.

Since we are brainstorming here, discussion is inappropriate. We will just get the ideas down then we can discuss them in the next step.


This next step allows us to discuss and clarify our brainstorming ideas. Before we can make decisions about which of these ideas we think are most important we have to be sure we understand each notion. Does anyone need clarification or want to talk about the logic behind their ideas? Let's start by weeding out duplicate ideas.


Please take some time to select the 6 items you think are most important and rate each #s 1-6 with 1 being the most important and so forth. This is a secret vote so, when you have made your choices, please /MSG JOYCE_ with your response.


Now is the time to discuss briefly, the results of the preliminary vote. We do this not so much for consensus, but for clarity. We don't have to agree we just have to be sure we all understand each other.


The final vote provides an opportunity to rethink and to codify the best thoughts of the group as a whole. This time I would ask that you make your vote a part of the record by typing them in one long comment to the channel like this... "1. blahblahblah, 2. etcetcetc,..."

Appendix C - Sample Demographics

sample size     =       27              16 male, 11 female

mean age        =       43.59           range 18 to 62 years

Tally for the following question:

"Please indicate the your level of experience with computer-mediated


novice                   intermediate                  experienced

(3) (2) (4) (4) (5) (9)

Appendix D - Tally of the final Vote for Nominal Groups

Sunday, March 24, 1996 (MOO) Wednesday, April 10, 1996 (IRC)
1. choice of words [1.56] 1. emoticons [2]
2. emote verbs [2.2] 2. asterisks [2.2]
3. use of punctuation [3.2] 3. actions [2.8]
4. pretending to do real things together [4] 4. abbreviations [3.8]
5. messages [4.2] 5. caps [4.6]
6. capitals [5.3] 6. metaphors [4.8]
Sunday, March 31, 1996 (MOO) Saturday, April 6, 1996 (IRC)
1. emotes [1.33] 1. emoticons [1.75]
2. tone [3.16] 2. words in * [2.75]
3. verbs [3.67] 3. hugs [3.75]
4. specific text [3.67] 3. emotionally charged phrases [3.75]
5. directed says [4.33] 5. actions [4]
6. commanding statements [4.83] 6. humor [5]
*Numbers in [ ] represent the numerical ranking average from the final vote in each nominal group.
Cumulative percentages were arrived at considering both frequency and rank order.

Cumulative Results
emotes (verbs or actions) [25% of total votes - mean = 2.95]
choice of words [25% of total votes - mean = 3.67]
use of punctuation [12.5% of total votes - mean = 2.72]
emoticons [8% of total votes - mean = 1.875 ]
use of capitals [8% of total votes - mean = 4.95]
miscellaneous [21% of total votes]

Joyce Menges ( is an online educator and a long-time participant in computer-mediated communication. In her face-to-face life, she teaches mathematics for Southern Maine Technical College and coordinates SMTC's NovaNET Learning Lab serving underprepared freshman. This research served as the senior theme for her master's degree in Adult Education awarded in May 1996 by the University of Southern Maine.

Copyright © 1996 by Joyce Menges. All Rights Reserved.

Contents Archive Sponsors Studies Contact