Developmental language learning impairments (LLI) are one of the most prevalent of all developmental disabilities, can occur in children for a wide variety of reasons, and have been shown to co-occur frequently with other developmental social, emotional and behavioral disorders, as well as with academic achievement problems. Research pertaining to developmental LLI of unknown origin, with an emphasis on the continuum between oral and written language impairment, is the focus of this review. Given the complexity of language learning, research has focused on multiple levels of analysis, including linguistic, neuropsychological, genetic, neurobiological, and remediation studies. To date, the vast majority of data on LLI derive from studies focused on a single level of analysis. Although attempts have been made to integrate data across studies and multiple levels of analysis, this has proven to be problematic, given the heterogeneity of the subject populations used to study LLI, as well as the differences in ages, degree of impairment, and types of impairment included in each study. Given that LLI is a complex developmental disability, it is suggested that future research would benefit from taking a multiple levels of analysis approach with the same individuals, incorporating mathematical models designed to analyze dynamically changing complex systems, and studying individual differences in language learning, prospectively and longitudinally, throughout the most dynamic stages of the process.