Case marked pronouns in English have received an increasing amount of attention from child language researchers from a Constructivist or Usage-based perspective (Tomasello, 2000; 2003; Kirjavainen et al., 2009) and also from a Generativist perspective (Schutze & Wexler, 2000; Rispoli 1998). However, genitive-for-nominative errors specifically, have received less attention. A detailed Corpus Analysis investigated English-speaking children's my-for-I pronoun case marking errors. The results revealed no direct link between frequency of my in the input either independently or as part of a larger construction and children's my-for-I error rates. However, the modelling of the pronominal system was found to be related to children's error rates. Children receiving mostly pronouns in the input, in contrast to a high proportion of proper name replacement, were found to have a lower rate of errors. Further, children with a less entrenched use of I, before their error period, were those who made a higher my-for-I error rate. Further analysis revealed an association between the function of agency, control and possession in children's erroneous uses of my as subject in verb phrases.An experimental study with two and a half and three and half year old children found that a range of pronoun case marking errors were being made alongside the correct use of I in both age groups. No age differences were found for the rate of correct I-verb utterances. This shows that children do not go through a period of time 100% making a certain error type and then change to 100% correct I, but that actually both forms are accessible and will compete for use. Within the study the claiming of agency and control was in focus, which elicited a high level of protest from the children. Analysis revealed that the children as young as two and a half were capable of protesting normatively, making use of normative language to show their awareness of the game rules and how things "should" be done. Further, developmental differences showed that the younger children may not be as sophisticated as the older children in altering their protest, according to context. This development trend shows that children may be learning more linguistic forms, but their understanding of when it is necessary to use them, may not yet be as advanced, in two year old children.This current thesis supports a Constructivist or Usage-based perspective of language acquisition. The importance of input, function and competition between forms has been exemplified. If children assign a form-function mapping between agency, control and possession and the form my, they are likely to make a higher level of error in these contexts. This error rate is then increased if the modelling of the pronominal system is low and not offering a high level of correct use of my and I, to compete with this erroneous mapping. This lack of modelling plus a lower entrenched use of I from the outset, will mean this competition will therefore be won out more often and for longer by this incorrect my form.