The major reasons for developing human monoclonal antibodies were to be able to efficiently manipulate their effector functions while avoiding immunogenicity seen with rodent antibodies. Those effector functions involve interactions with the complement system and naturally occurring Fc receptors on diverse blood white cells. Antibody immunogenicity results from the degree to which the host immune system can recognize and react to these therapeutic agents. Thus far, there is still no generally applicable technology guaranteed to render therapeutic antibodies antigenically silent. This is not to say that the task is impossible, but rather that we need to train the immune system to help us. This can be achieved if we take advantage of natural mechanisms by which an individual can be rendered tolerant of "foreign" antigens, and as a corollary minimize the potential immunogenicity of any contaminating protein aggregates, or "aggregates" arising from antibodies complexing with their antigen. I here summarize our efforts to engineer antibodies to harness optimal effector functions, while also minimizing their immunogenicity. Potential avenues to achieve the latte are predicted from classical work showing that monomeric "foreign" immunoglobulins are good tolerogens, while aggregates of immunoglobulins ate intrinsically immunogenic. Consequently, I argue that one solution to the immunogenicity problem lies in ensuring a temporal quantitative advantage of tolerogenic non-cell-boundmonomer over the cell-binding immunogenic form.