169centralised around Ego, who is able to master his or her connections
because alters have no direct connection with each other. Individuals with
bridging social capital have distinct subsets of network members, for
which they may become bridgers or brokers; they do not benefit, however,
from a cohesive group of network members who know or often interact
with each other. On the one hand, individuals with only bonding social
capital available in their personal networks have interconnected support-
ive alters who may help them to progress in their local environment. On
the other hand, this redundancy of ties impedes their contact with a
diversity of world views or information and to some extent jeopardises
their autonomy.
Despite these two understandings, bonding and bridging social capitals
should not be seen as mutually exclusive. Individuals who have a compre-
hensive type of social capital, both bridging and bonding, are likely to be
the best adapted to the constraints of modern life, as they benefit at the
same time from collective support if needed and from having set foot in
distinct relational worlds (Widmer 2006 ). However, a brokerage position
requires a personal investment in time, energy, and sociability necessary to
create and maintain autonomous personal connections. Therefore, it is
not easy to gain and to hold bridging social capital, and much of it
depends on the other resources (economic, time use related, emotional)
that individuals can invest in relationships with their alters.
In summary, bonding and bridging social capitals should not be inter-
preted as two opposite types of social capital as they may, in some cases,
occur together. Various measures should therefore be used to operation-
alise bonding and bridging social capital made available by personal net-
works. First, the sheer number of individuals present in a personal
network is an indicator of social capital: the more members there are, the
larger the social capital, as greater size means more heterogeneity of com-
position and weaker density. However, other measures beyond size are
needed, as not all members of a personal network might be activated in
all kinds of support. Density, which is computed as the number of exist-
ing connections divided by the number of pairs of network members
cited by Ego and corresponds to potential connections, is indeed a central
indicator of bonding social capital. In one of the founding articles on
social capital, James Coleman ( 1988 ) stressed the beneficial impact of
Understanding Personal Networks as Social Capital