research, he concludes that most accounts of virtual community are
over-inflated, and that this has led to a simplistic stereotyping of most Net
users as isolated nerds and geeks who live out a withdrawn and isolated
existence. As I have argued above, Castells’ claims are based on the col-
lapsing of demographic surveys which set out from an institutional
framework with research into CMC. This does not leave him with any
room for a considered analysis of category 3. Either such a category is
reserved for a tiny minority of persons with identity problems who seek
the Net as some sort of fantasy or refuge from real life, or else Castells is
anxious to show that such persons also have a life off-line where they lead
quite normal existences, and are ‘civically engaged’; i.e. that there is no
withdrawal from other forms of sociability. This again shows that, for
Castells, the ‘social’ is a decisively off-line condition and that life on-line
either enhances or annuls such sociability.
It is true that, where there is total overlap between the two kinds
of communities, we typically think of on-line communication as an augmen-
tation or supplementation of face-to-face community rather than the other
way around. In the case of partial overlap, the moment of identification
between two communicants is often more ambiguous and difficult to
negotiate. For virtual communities which have no reference to physical
communities, identification is not burdened with ‘verification’ in that
identity formation is purely internal to the speech events.
Because of this, there is a sense in which, in the third situation, part
of every communication event consists, in part, of interlocutors’ conver-
sations with themselves. Taylor argues that in CMC, ‘The striving subject
enters into conversation in order to build itself up through the search for
truth. Thus the person who converses relates to herself/himself even
when s/he seems to be relating to an other’ (cited in Foster, 1997: 26). The
computer terminal is less a window onto other worlds than it is a half-
reflective mirror. Psychoanalysts would tell us that this is true of all
communication, but in cases where such communication becomes tech-
nologically embedded, it also gains a different kind of visibility. Where
there are low levels of identification, the case of Aoki’s third form of com-
munity, the validation of truth, morality and aesthetics, the three norma-
tive domains inherited from modernity cannot be satisfied dialogically,
except in an imaginary sense, an illusion that is proportional to the
accountability of each communicant.
But the low accountability of on-line communicants, what I have called
‘interaction without reciprocity’, isn’t merely related to the absence of
identification between interlocutors but also to their high virtual motility –
the potential for mobility, including the mobility to ‘switch off’ and escape
from any communication event. This feature is accentuated by the high
number of possible connections which can be made on-line, and the fact
that such connections are part of a web or a network.
Rafaeli and Sudweeks (1997) have argued that on-line interactivity
needs to be thought of across an entire network, not simply betweenTelecommunity 205Holmes-06.qxd 2/15/2005 1:03 PM Page 205