Willingness to communicate in English is an “umbrella” concept encompassing a variety of individual factors which, when combined, describe a readiness to speak in the target language. Whether Willingness to Communicate (WTC) is a permanent trait or is modified by situational context has previously been investigated in various studies. More recently, the importance of using dynamic systems theory to describe the complex and interrelated properties of WTC has been revealed (MacIntyre & Legatto, 2011). Moreover, the nested ecosystems model (van Lier, 2002) may be one means of describing a dynamic operating system such as WTC. A few WTC researchers have endeavoured to situate their subjects within such nested systems, in an effort to take full account of the various layers of context by which learners are influenced and affected. However, a combination of both these theoretical approaches has yet to be fully explored in the WTC research field. As a result, this study used mostly qualitative methods to obtain data from ten adult Iranian migrants studying English in New Zealand, as well as their ten classroom teachers, in order to exemplify the nature of this dynamic system (WTC). Semi-structured and stimulated recall interviews were conducted several times with the participants and their teachers, after observations had taken place in their classrooms and a questionnaire completed. In order to create a more detailed picture of the participants’ WTC, it was designed as a longitudinal study carried out over a period of eighteen months and included an investigation of variables affecting WTC from outside the classroom, as well as past learning experience. Findings suggest that a range of variables influenced the classroom WTC of these learners, both external factors (out of the students’ control), such as the teacher, texts and methods used, and class activities, as well as internal factors, such as self-perceived English-speaking competence, confidence, anxiety, motivation, and personality. Such antecedents were found to create individually different dynamic fluctuations in the levels of participants’ WTC. Thus, the inherent variability yet interconnectedness of a dynamic system was highlighted. In addition, two individual case studies, represented in an ecosystems framework, revealed the way in which various factors influenced their WTC through the porous nature of the four layers (i.e., micro-, meso-, exo-, and macro-). Therefore, the contribution of this study of WTC is to explain how Dynamic Systems Theory can provide an ecological perspective, and thus more fully contextualise language learning and use, particularly for migrants. It extends the pyramid model of MacIntyre et al. (1998) to include an element of time as supported by ecosystems theory. It expands the scope of research to WTC outside the classroom and finally suggests a new extended definition of WTC to guide other researchers in the future. Moreover, by revealing how contextual factors combine to enhance or inhibit students’ WTC, the findings of this study encourage language teachers to understand their students as holistically as possible and to accommodate variations in their daily WTC, which could be the result of both in- and out-of-class influences. Such investigations into the WTC of L2 students, especially new migrants, are warranted, as, globally, migration has become a very prominent issue in many countries.