This dissertation measured the acoustic properties of the English fricatives and affricates produced by native and Chinese L2 speakers of English to identify the phonetic basis and sources of a foreign accent and to explore the mechanism involved in L2 speech production and L2 phonological acquisition at the segmental level. Based on a Network Model of L2 Speech Production, it is hypothesized that L2 speech differs from native production substantially and these differences arise from the constraints imposed on the process of L2 speech production and from the native and non-native speakers. different mental representations of the segments in question. Three experiments were conducted to test the hypothesis at two levels, contrastive (phonemic variants) and non-contrastive, including syllable position conditioned allophonic variants and speech context conditioned contextual variants. Supporting evidence was found in the experiments that demonstrates that 1) the fricatives and affricates produced by Chinese L2 speakers differed from the native production substantially due to the multiple interacting constraints in L2 speech production; and 2) the fricative and affricates produced by Chinese L2 speakers did not vary significantly as a function of syllable position and speech contexts, suggesting a difference in the mental representation of the segments between the two groups of speaker. It is possible that phoneme is an abstract concept for L1 speakers, but a concrete instantiation of an L2 segment for L2 speakers or at least a much more impoverished representation, one that does not include nearly as much context-dependent variability. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]