The focus of this study is the relationship between language attitude, beliefs, efficacy, English language competence, and language achievement. Two hundred and eighty-five students from five metropolitan primary schools in Western Australia completed a specially designed questionnaire based on the Attitude/Motivation Test Battery (Gardner, 1985a). A different language was taught in each of the five schools: French (24%), German (19%), Indonesian (19%), Italian (16%), and Japanese (22%). Fifteen percent of students spoke another language in addition to English. Thirteen percent of students were not born in Australia. No student had been in Australia for less than two years. After controlling for the effects of gender, age, and language studied, language efficacy was found to be a significant predictor of language achievement. However, this effect disappeared in the presence of English-speaking competence, which remained as the only significant predictor from the group of language affect and English competency variables. A confounding effect was observed for the variable language studied. These results tend to support Cummins' (1992) suggestion that first and second language proficiency have a common underlying interdependence. Given these tentative findings, the research reported in this paper shows that the interrelationships between affective factors and language learning are both complex and dynamic and ripe for further exploration. (Contains 4 tables.)