1The role of morphological theories in understanding how the lexicon is organized has been the topic of research of a number of studies. This topic brings two contrasting views to the fore. Some researchers claim that the lexicon consists of independent polymorphemic words with no need of a decomposition theory. This approach has been advocated in Amazigh and other Semitic languages like Hebrew and Arabic, pointing to the fact that the morphology of the language is better explained using a word-based approach. Arguments have been presented from Hebrew verb morphology (Bat-El, 1994; Aronoff, 1994; Ussishkin, 1999) and from Arabic verbal and nominal morphology (Hammond, 1988; McCarthy and Prince, 1990; Guerssel and Lowenstamm, 1996; Benmamoun, 2003).

2Contrastively, others argue that polymorphemic words are decomposed into morphemes among which we cite the base form or the root (Cantineau, 1950; McCarthy, 1981; Galand, 1984; Chaker, 1990; Tobin, 1990; among others). Evidence for the root-based theory has been provided from language games (Arabic: McCarthy, 1981; Tashlhit: Lahrouchi, 2004, 2018a), metathesis (Prunet, Béland and Idrissi, 2000), and from behavioral studies (Deutsch, Frost and Forster 1997, 1998, 2000; Boudelaa and Marslen-Wilson, 2001, 2004a-b, 2005; Ussishkin and Twist, 2009; Ussishkin, Dawson, Wedel and Schluter, 2015). According to this view, the root is accessed very quickly in studies of language processing.

3In the present work, we contributed to the debate on the two views on morphological theory and discussed the theoretical implications for the organization of the lexicon. We tried to investigate the notion of roots in Amazigh, more particularly in Tashlhit and we attempted to answer two main research questions. First, is the root a morphological unit in the Tashlhit lexicon? Second, is the root exclusively consonantal in Tashlhit? With this end in view, we investigated the lexical properties of root structure in Tashlhit by distinguishing between two types of roots, the vocalic and the consonantal. We provided arguments supporting the claim that in addition to consonantal roots, the Tashlhit lexicon consists of roots that have vowels and consonants alike.

4The novelty of this dissertation is that it discusses the significance of the root from a perspective that is not only purely morphological but also psycholinguistic. We presented further arguments for the presence of vowels in Tashlhit roots and also conducted, for the first time in Amazigh studies, priming experiments to examine language processing. The data analyzed comes essentially from the variety spoken in Ighrem N’Ougdal area and its surroundings, a place which is largely representative of the Tashlhit language.

5At the theoretical level, we carried out our analysis under the premises of Optimality Theory (Prince and Smolensky, 1993/2004; McCarthy and Prince, 1993, 1995). We presented facts from the verbal and nominal morphology of the language, and we tried to account for the linguistic irregularities through constraint ranking. We resorted to Correspondence Theory (McCarthy and Prince, 1995) and to Positional Faithfulness theory (Beckman, 1998) with main reference to the root morpheme to account for some aspects of the morphological system of the language. Using the interaction between faithfulness and markedness constraints and root faithfulness constraints, we also argued for the presence of vowels in the root structure of Tashlhit assuming that input root elements, be they vocalic or consonantal, are preserved in the output.

6At the psycholinguistic level, we followed the assumption that linguistic phenomena are not exempt from extralinguistic factors (Berent & Shimron, 1997, 2003; Frish & Zawaydeh, 2001; Prunet, Béland & Idrissi, 2000). More specifically, we discussed data from priming experiments, based on measuring the reaction times of the participants. This type of experiment generally exposes participants to a stimulus (prime), which influences their response to a subsequent stimulus (target).

7This dissertation is organized into five chapters. Chapter I sketches the characteristics of the Tashlhit phonological and morphological system, on the one hand, and the root structure in Hamitic-Semitic languages, on the other. Chapter II presents the premises on which our analyses are based. As a reminder, our work is organized in a twofold fashion: one part provides the theoretical analysis and the argumentation for the root structure and the second part adduces external evidence for the significance of the root. In this chapter, we, first, present an overview of the Optimality Theoretic framework, Correspondence Theory and Positional Faithfulness. Second, we introduce the priming tests we used in conducting our experiments. In chapter III, we provide further arguments in favor of the root-based approach. We discuss the relevance of the root in understanding the verbal and nominal morphology of Tashlhitin further detail, and we also distinguish between vocalic and consonantal roots in the Tashlhit lexicon. Ample evidence is presented in support of the bipartite system of Tashlhit roots. In chapter IV, we test the semantic priming effect in the language as we have more regular overlappings of semantics and morphology in the language. We present the details of the methodology adopted in all the priming experiments we usedin this study. We also introduce the pretests we used to select our experimental stimuli, namely the frequency and the semantic pretests. Last but not least, chapter V examines the morphological priming effect as external evidence for the role of morphology in the language and tests the other linguistic factors (semantics and phonology) that may interfere with the root.

8The results of our theoretical and empirical analyses showed that the root is an essential morphemic unit that plays an important role in the understanding of language processing. We proved that roots in Tashlhit have some psycholinguistic reality and, hence, they have significant implications for the organization of the Tashlhit lexicon. We obtained the same result with semantic features that showed a significant priming effect, suggesting the lexicality of semantic features in the Tashlhit lexicon. Only phonological properties did not facilitate lexical access, leading to the conclusion that phonology has no role in word recognition processes. We also argued for the coexistence of both consonantal and vocalic roots in the Tashlhit lexicon and that the vocalic element in the root morpheme is not positionconstrained but, rather, can occupy the initial, medial and or final position of the root.