including computers.  Virtual worlds exist not in the technology usedto represent them, nor purely in the mind of the user, but in therelationship between internal mental constructs and technologicallygenerated representations of these constructs.  The illusion ofreality lies not in the machinery itself, but in the users'willingness to treat the manifestation of their imaginings as if theywere real.The technical attributes of these virtual places, comments Curtis,have significant effects on social phenomena, leading to new modes ofinteraction and new cultural formations.[7]  The lack of actualphysical presence, indeed the great physical distances betweenindividual participants, demands that a new set of behavioural codesbe invented if the participants in such systems are to make sense toone another.  The problems posed by the lack of cultural cues whichphysical presence carries influence behaviour in virtual environments.The solutions to these problems which participants devise constitutethe culture of the virtual world in which they are played out.  It isthe tension between the manifestation of conventional social andcultural patterns, the invention of new patterns, and the imaginativeexperience of these phenomena as taking part in a virtual world thatis the subject of my thesis.My primary sources in this work fall into three categories.  Firstly,I will quote from logs taken of sessions on MUDs.  Secondly, I willquote from electronic mail, or email, sent to me by MUD players inwhich they discuss such usage.  Lastly, I will be using articles fromthe USENET newsgroups devoted to discussion of MUD and MUD playing.These groups include alt.mud, rec.games.mud, rec.games.mud.admin,rec.games.mud.announce, rec.games.mud.diku, rec.games.mud.lp,rec.games.mud.misc and rec.games.mud.tiny.  I have been monitoringthese groups since December 1991, during which time these groups haveseen an average traffic of approximately fifty articles each day.  Inall quoted extracts the original (sometimes very original) grammar andspelling have been preserved, and in all cases I have securedpermission to quote from the individuals concerned.  In some cases Ihave been asked to withhold identifying information, and where this isthe case I have indicated in the footnotes that the item of mail orthe news article is from "anonymous".  However, in most cases thenames of players and characters as well as the names of the MUDsthemselves have been preserved.  The most important exception is thecase of 'JennyMUSH', which is an alias.  For reasons that will bemade clear in the body of this thesis, the unique nature of thissystem and the experiences of its users have led to a great concernwith the issue of privacy.  The administrator of the MUD has asked menot to reveal any information that might identify the location of thesystem, and has suggested 'JennyMUSH' as a pseudonym which retainsthe flavour of its actual name.This thesis will be divided into three chapters, preceded by a sectiondetailing the historical background to and context of the evolution ofMUD systems.  The subject of the first and second chapters is thenature of the social changes that these forms of virtual realityengender.  I will examine the impact of MUDs on the practices ofinterpersonal communication and interaction, and on communityformation and social cohesion.  The third chapter will describe howthe nature of human existence is altered by entrance or translationinto virtual reality.  In this last chapter I will explore the natureof social identity, sexuality and the body in the virtual environment.---FOOTNOTES TO INTRODUCTION---[1]  William Gibson, _Neuromancer_ (London: Grafton Books, 1989) 67.[2]  Nicholas Lavroff, _Virtual_Reality_Playhouse_ (Corte Madera CA:     Waite Group Press, 1992) 7.[3]  Michael, Benedikt, _Cyberspace:_First_Steps_ Cambridge,     Massachusetts: The MIT Press, 1991) 1.[4]  Howard Rheingold, _Virtual_Reality_, (London: Mandarin, 1992) 46.[5]  Pavel Curtis, "Mudding: Social Phenomena in Text-Based Virtual     Realities,"  _Intertek_ Vol. 3.3 (Winter, 1992) 26.[6]  Curtis, 26.[7]  Curtis, 26.------------------------------------------------------------Background: A History of Interactive and Networked Computing------------------------------------------------------------and the Evolution of MUDs----------------------------INTERACTIVE COMPUTING[1]---Personal computers are a relatively recent phenomenon.  It is onlywithin the last ten to twenty years that such machines have becomecommon in the work place, let alone the home.  The pre-history ofcomputing was largely the domain of educational, governmental orcommercial organisations which owned large mainframe computer systems.These huge old systems were jealousy protected; computer time washeavily booked and access available only to the privileged few.  Thesecomputers of the past generation would hardly be recognisable to thepresent generation of Mac and PC users.  The old beasts of the '50sand '60s took up literally rooms of space.  Their computing power wasmeasured not in millions of instructions per second--MIPS--but inhundreds of instructions.  The multiple megabytes of random accessmemory we now take for granted in even the most humble of desktopsystems were then only a fantastic dream.  The greatest and mostcostly super-computers of the sixties counted their memory inkilobytes, hard and floppy disks were yet to be invented, monitors andkeyboards were only in the experimental stages, and most computersreceived instructions and gave back results on long spools of punchedpaper tape.Still, archaic as these clumping monsters now appear to be, they werethe gleaming prize of their age.  Mathematicians, statisticians,physicists, military engineers and government agencies all fought forthe funding to acquire one of these miraculous new machines.  Theyalso attracted the interest of a new breed of young inquiring minds.At the Massachusetts Institute of Technology, one of the feweducational institutions to invest large sums in the new computingtechnology, the members of the Tech Model Railroad Club switched theirinterest from the construction of intricate train tracks to themanipulation of complex computer circuits.[2]  Of course these youngstudents, most of them undergraduates, were not able to get directaccess to the new machines.  Instead they took to hanging around thecomputer rooms at midnight and the small hours of the mornings,begging computer time from the nightwatchmen on the few occasions whenthese least attractive hours had not been booked by others.Most of the computers of the time relied on punched paper both toreceive instructions and to communicate results.  This forced computerprogrammers and users to divide the giving and receiving of data intodiscrete blocks.  Instructions would be transcribed into the punchedcode useable by the computer, the instructions would be acted on bythe computer and the results of its computations spat back on punchedtape.  These results would then have to be decoded before any furtherwork could be done.  MIT's academics--physicists and statisticiansand mathematicians--relied on and accepted this paradigm of computeruse.  Not so the members of the Tech Model Railroad Club.  Theirinterest quickly centred on an experimental computer which the DigitalEquipment Corporation had loaned to the Institute.  This computer wasmuch less powerful than its hulking IBM cousins, and so was virtuallyignored by the academics to whom it had been lent.  It was adopted bythe TMRC students because it offered a new paradigm of computing.DEC's Programmed Data Processor was among the first to incorporate ascreen and a keyboard.The TMRC members had no complex scientific problems to solve.  Insteadthey spent their time simply exploring the capabilities of the PDPmachine.  They programmed to demonstrate their skill in understandinghow the machine 'thought'.  Staying up all night, and functioning,so the story goes, on a diet of coke and burgers, these young'hackers' set out to colonise the unexplored territory of thecomputer.  One of their most famous endeavours was the invention ofthe first computer game.  By modern standards it was uncomplicated.  Asimple figure of a spaceship appeared on the screen, to be shot downby the player.  At the time, however, it was a marvellous feat ofcomputer graphics, a miracle of programming.  Copies of  'Spacewar',in punched paper form, were passed around to computer enthusiasts atother institutions, and began a small revolution in computer use.[3]The game of Spacewar depended on human/computer interactivity.  Itrelied on the human user being able to monitor the computer's actionsand modify and correct for them while the machine was actuallyoperating.  The concept of human/computer interaction did not beginwith this invention of the computer game, but the game made a smallinstance of this interactivity available to a rapidly expanding numberof computer users and demonstrated that such concepts could berealised in a simple and 'user-friendly' fashion.  It brought newprogramming ideas--new algorithms--to the computing world.  It alsochanged the way that the academy thought about computers.  The leapbetween the idea of computers as awesome inhabitants of super-cooledrooms, tended by white-coated engineers, to the idea of the computeras toy and expressive tool, was made when that first spaceship wasshot down.  Spacewar made tangible the idea of the computer as amedium for human expression.---NETWORKED COMPUTING[4]---The computing expertise of the TMRC members soon came to the attentionof MIT's authorities.  Wishing to harness this obvious talent, MITgave the students legitimate access to the computers, and legitimatework to perform on them.  One of the first jobs they were assigned wasto solve the problem of the costs involved in buying enough computersto cater for the increasing numbers of people who wished to use them.MIT was considering investing in a new form of operating system, knownas the Compatible Time-Sharing System, which would allow more than oneperson to use a computer at once.  Instead, in a cost-saving move,they set the TMRC students to designing their own multi-user operatingsystem.  The multi-user computer system relied on a different hardwareto the single user system.  If more than one were to be accommodated,there needed to be more than one set of input and output devicesconnected to the computer.  From each of these multiple terminals,different users could share the same computer resources.  The systemthat they designed, and named the Incompatible Timesharing System, wasone of the first of this new breed of operating system.  ITS and othersystems like it quickly supplanted the old single-user systems.Today, the most popular multi-user operating systems are part of theUNIX family, descendants of a system which Bell Laboratories began todevelop in 1969.The multi-user paradigm quickly became popular, as its cost-effectiveness became apparent, and was followed by the idea of thecomputer network.  Programmers in the United States Department ofDefence built the first network.  In 1969 the DoD began work on a'long-haul' network of computers at dispersed sites.  This projectwas funded by the Advanced Research Projects Agency, a research arm ofthe DoD.  The original purpose of the ARPANET project was to design asystem for use by military control and intelligence.  The network wasdesigned to enable authorities to communicate and weapons to becontrolled remotely in the event of a nuclear war.  The problem withwhich the engineers who designed the system were faced was that duringa war any central control point would most likely be the target ofenemy missiles.  The solution was a network structure that had nocentral point and which was designed from the beginning to withstandphysical attack.  Each node of the network could operate as a centralpoint, and there would be no 'right' way for a message to bedirected from one node to another.  Messages could follow any route,and should one node be taken out of operation, messages would simplyskirt around it.  This rather haphazard delivery system would beextremely resilient--even with large portions of the network knockedout, information could still be transmitted.[5]In 1969 ARPA set about installing the first node of the network at theLos Angeles campus of the University of California.  Shortly afterwardnodes were installed at the Santa Barbara campus of the sameuniversity, at the University of Utah, and at the Stanford ResearchInstitute.  Once the system was up and running, these universitieswere given leave to use it for research purposes.  They jumped to doso, planning to exploit the network's ability to give users of thecomputers at each of these sites access to the resources held by allthree.  At the same time, DARPA encouraged other institutions to setup their own network nodes, each of which could be commandeered intime of war.  By 1972 thirty-seven universities and governmentresearch organisations were on ARPANET, and as the network grew theseinstitutions began to demand autonomy from the military.  In 1983ARPANET was divided into two networks, known as ARPANET (for researchuse) and MILNET (for military use).  The ARPANET arm continued toexpand, with local area networks at various government, educationaland commercial sites being added to the system.  Other nations alsoadopted the technology, and with the advent of satellitecommunications, it became possible for all these computer networks tobe linked together as one super network.  This new internationalentity became known as the Internet.