It is striking that such different and largely uncorrelated temperament styles both would be associated with adolescent alcohol problems. Multiple pathways to alcohol problems are usually discussed in the context of internalizing and externalizing pathways (Zucker et al., 2008). However, our findings suggest a more nuanced understanding of different pathways to alcohol problems when starting in very early childhood. We do not find a clear externalizing/internalizing distinction in our empirical evaluation of very early childhood temperament, with both emotionality and other indices of conduct difficulties as indexed before the age of 5 all loading together. A further point is that the other temperament style associated with alcohol problems in the ALSPAC sample (sociability) is not one that is on the surface maladaptive. In studying risk factors for adolescent alcohol consumption and problems, it is important to remember that alcohol use is largely a social phenomenon in adolescence. Experimentation in adolescence is normative. More than a decade ago, a longitudinal study of drug use among adolescents followed from preschool through age 18 (Shedler and Block, 1990) showed that adolescents who had
