Deixis and EFL reading comprehension.

Abstract

The purpose of this research is to determine the effect of deixis as pragmatic residues on EFL reading comprehension and to find out whether the injection of deictic expressions in reading passages would jeopardize or hamper the comprehension of the readers at all different levels of language proficiency.

Introduction

Reading is a complex activity which involves identification of letters, recognition of words, visual discrimination and thereby sampling of ideas, predicting the forthcoming content and decoding the intended meaning of the message. (Varzegar, 1993) Reading is a process of idea construction, confirmation, rejection and semantic interpretation.(Varzegar, 1993). Varzegar(2003)asserts that reading is a process of recognition of words and structures while associating meaning with them and understanding the contextual, conceptual, scriptual, schematic and pragmatic meanings which lead to a more global and holistic comprehension. Comprehension of a text involves seeing the interrelationships, intrarelationships and suprarelationships of sentences, paragraphs, the organization of longer passages and many other factors. The reader may read sentence by sentence and comprehend the individual sentences, but fail to extract the meaning of the text because he does not grasp the pragmatic residues (Levinson, 1983), such as deictic expressions in the passage. Moreover, the ultimate goal of reading is comprehension and the optimal level of comprehension is "demessaging" (Carton, 1976) the printed page. Varzegar (2001) maintains that demessaging goes beyond the linguistic aspects of the text and brings the extralinguitic, discoursal, scriptual, schematic and pragmatic knowledge into the reading act.

Furthermore, there are two basic processes involved in comprehension: (1) decoding and (2) demessaging. "Decoding is the process of trying to understand the meaning of a word, phrase, or sentence" (Pratt and Weber, 1985, p. 73) and demessaging is the scriptual, schematic and pragmatic aspects of comprehension" (Varzegar, 1995, p.112). Carton(1976) believes that the message cannot be truly comprehended unless it is demessaged. Barnitz (1985) sums up three essential elements of an adequate

model reading. First, reading is "multileveled" in that readers use of various levels of language simultaneously to access meaning. Readers use their "pragmatic," discoursal, syntactic, morphological and phonological knowledge in constructing and reconstructing meaning. Second, reading is "interactive," in that the reader's comprehension is driven by the knowledge structures or schemata of the reader and the specific content and linguistic structures in the text. All levels of background knowledge (linguistic, social, conceptual)interact simultaneously as the reader extracts meaning from the text. Third, reading involves the generation of hypotheses as the reader makes predictions about the meaning of a text.

To conclude, one should take into account that reading is an active, interactive (Rumelhart, 1979),transactional (Rosenbalt, 1994), interactive-compensatory (Stanovich, 1980) process in which not only the reader's background knowledge (schema) interacts with what is on the printed page, but also different levels of his/her competence; that is, linguistic, communicative, schematic and pragmatic competencies interact with each other. There also exists an interaction between all sub-components of language, such as phonology, morphology, syntax, semantics and super components of language, such as discourse and pragmatics, as well as an interaction between all other language skills. The modern usage of the term PRAGMATICS is attributable to the philosopher Charles Morris (1938), who was concerned to outline the general shape of a science of signs, or semiotics. Pragmatics is the study of "the relation of signs to interpreters (Levinson, 1983). Carnap (1938) defines pragmatics as such: "If in an investigation explicit reference is made to the speaker, or to put it in more general terms, to the user of the language, then we assign it to the field of pragmatics."

Levinson(1983) amends Camap's definition to "Those linguistic investigations that make necessary reference to aspects of the context." Pragmatics covers both context-dependent aspects of language structure and principles of language usage as well as language use. "A sentence is an abstract theoretical entity defined within a theory of grammar, while an utterance is an issuance of a sentence, a sentence-analogue, or sentence-fragment, in an actual situation (context)." "Pragmatics is concerned with the study of meaning as communicated by a listener (or reader).... Pragmatics is the study of contextual meaning." (Yule, 1997) Morris (1938) defines pragmatics as the aspects of language that involve users and contexts of use of linguistic expressions (cited by Green, (1996). Bar-Hillel(1954) proposed that "pragmatics be concerned with indexical expressions, whose (1955) meanings can only be determined relative to user and context of use. "Consider the following examples:

   They are here.
   I saw you there yesterday.
   After that thing happened.

The single most obvious way in which the relationship between language and context is reflected in the structures of languages themselves, is through the phenomenon of deixis. The term is borrowed from the Greek word for pointing or indicating, and has as prototypical or focal exemplars the use of demonstratives, first, second person pronouns, tense, specific time and place adverbs like now and here. Longman Dictionary of Applied Linguistics (1996) defines deixis as "A term for a word or phrase which directly relates an utterance to a time, place, or person. Deixis is defined by Fillmore (1966) as "the name given to those aspects of language whose interpretation is relative to the occasion of utterance." Harman (1990), in the same line, believes that any words or phrases whose referents depend on the immediate situation of utterance, n terms of participants (speaker/hearer), place, and time are said to be deictic. Consider the following example:

Meet me here a week from today with a stick this big.

The words "me," "here," "today," and "this," do not have clear referents and are very difficult to understand. Since no research has already been conducted on the effect of pragmatics and reading comprehension, a study was designed to investigate whether there is a significant difference between the reading comprehension of reading passages with and without deixis among the EFL learners with different levels of language proficiency.

Method

Subjects The subjects participating in this study were all Iranian undergraduate students majoring in English literature and English translation. They outnumbered 162 students, all of whom came from two universities: University of Tehran and Azad University. Out of this number of subjects ho were sophomores and seniors and who were chosen randomly, 18 subjects were discarded in order to have equal sample sizes for a two-way ANOVA. The remaining 144 subjects were divided into three groups: elementary, intermediate and advanced, based on their performance on the Nelson-Denny Reading Test. The elementary group consisted of 32 subjects, the intermediate 82, and the advanced group 30 subjects. For the sake of equal size sampling, 52 subjects were discarded equally from the highest and the lowest status in the intermediate group. Two subjects were also discarded from the mid-point of the distribution of the elementary group. Therefore, the total number of subjects in each group was adjusted to 30. This sample population provided the main data for this experiment.

Instrumentation Two tests were utilized in this study: the Nelson-Denny Reading Test(NDRT) and the Reading Comprehension Test. However, the two tests were presented to the examinees as one test containing three sub-tests: Vocabulary, Comprehension (the two sub-tests of the NDRT) and Reading Comprehension.

The Nelson-Denny Reading Test The Nelson-Denny Reading Test was used in this study as a standard test of reading comprehension. This standard test was utilized to divide the subjects into different reading comprehension levels, since the focus of this study was on reading comprehension. The Nelson-Denny Reading Test is a well-known and trustworthy test developed for the purpose of testing reading comprehension ability. From among the different versions of the Nelson-Denny Reading Test available, the latest version, that is Form H, was used for this study. This form enjoys a high degree of test reliability (0.87).

The Reading Comprehension Test The second instrument of this study was the Reading Comprehension Test, which was attached to the NDRT (to test their true reading comprehension ability) and was introduced to the testees as the third sub-test of the NDRT. This test consisted of 6 reading comprehension passages selected from the standard Test of English as a Foreign Language (TOEFL). Two of the passages were selected for the elementary group with a readability of 13 (low), two other passages were selected for the intermediate group with a readability of 17 (mid) and the other two passages, with a readability of 22 (high), were selected for the advanced group. The passages were all narratives, since the purpose of this study was to load some of these passages with deictic expressions. Narratives provide an appropriate ground for manipulating the time, place, and person deixis. For each group, one passage was left intact and the other was loaded with time, place and person deixis. An effort was made by the researchers not to change any other elements of the passage except to replace the non-deictic terms with similar deietic ones. It should be mentioned that although the passages were manipulated and non-deictic terms were replaced by deictic ones, no exploitation was done on the comprehension items; that is, the comprehension items of the passages were left intact. Moreover, in order to motivate the subjects to answer the comprehension test and in order to eliminate any possibility of hint to the manipulation of the passages in the Comprehension Test, the six passages were retyped (exactly following the NDRT's format), reprinted and attached to the Nelson-Denny Reading Test, as if they ere he continuation of the seven reading comprehension passages of the NDRT. On the whole, the Reading Comprehension Test consisted of twenty-nine items and the examinees were given thirty minutes to answer the items.

Procedures

Prior to the study, a pre-test was given to 30 subjects who were randomly selected and were similar to the target group. The pre-test consisted of the originals of the six complete reading comprehension passages selected from the TOEFL Test. The purpose of this test was to determine whether the test benefits from an adequate degree of reliability. Moreover, prior to the actual administration of the tests a pilot study was conducted. Few subjects who were randomly selected received the test. Some of the subjects were instructed to answer the Reading Comprehension first, then the Vocabulary part and finally the Comprehension part; others were instructed to answer the Vocabulary first, then the Comprehension and finally the Reading Comprehension. The results indicated that the examinees would perform better if they had answered the Reading Comprehension Test first and the Nelson-Denny Reading Test next. Therefore, the same procedure was instructed to the target group. All the examinees received the Nelson-Denny Reading Test and the six comprehension passages from the TOEFL Test. The practice, however, was to correct only the passages that were related to each group after the results of the NDRT was determined. The scoring procedure for the

Nelson-Denny Reading Test was on the basis of giving one point to every correct answer of the Vocabulary section and two points to every correct answer of the Comprehension section. The reason for this double weight of the Comprehension items put forth by the authorities of the test was that the Vocabulary items were twice the number of the Comprehension items. The scoring of the Reading Comprehension Test followed a procedure of giving one point to every correct answer. There was no penalty for the wrong answers in the test.

Design

The design adopted for this study was an "EX Post Facto" design. This type of design is often used when the researchers look at the type or degree of relationship between the two variables rather than at a cause-and-effect relationship. Moreover, in such designs, the researchers have no control over what has already happened to the subjects. The treatment, whatever it might be, has been given prior to the research project. In such studies, since there is no casual relationship between the two variables, the distinction between independent variables and dependent variables is not well-defined. It is arbitrary to call one or the other the independent variable (Hatch and Farhady). Therefore, the researchers have also defined different types of variables in this study arbitrarily.

Data Analysis

Readability In order to be able to determine the appropriate readability of the passages for each level, 50 comprehension passages were randomly selected from the TOEFL Test. The reliability of each passage was computed through the Fog Index Formula. The mean and the standard deviation of these readabilities were also calculated as: Mean(X)=16.6 Standard Deviation (SD)=4.35 Then, the limits for the readability bands of z = -1 and z = +1 were calculated in order to determine the readability of passages for the elementary, intermediate and advanced groups. The readability stood for z of -1 at 13, excluding the 13, and for z of =1 at 21, excluding the 21. The readabilities within the defined range were considered as appropriate for the intermediate level; those below z of-1, appropriate for the elementary level; and those above the z of +l, appropriate for the advanced level.

Reliability As it was mentioned previously, the Reading Comprehension Test consisted of 29 multiple-choice items, with a total score of 29 as each correct item was weighed with one point. The original forms of all six passages that were selected from the TOEFL Test were administered to a total number of 30 subjects, similar to the target group of examinees, in order to obtain its reliability. The mean and the standard deviation of this administration were calculated as: Mean (X) = 11.27 Standard deviation (SD) =6.65 The KR21 Formula was utilized for the computation of the reliability coefficient of the Test. It was found to be: Reliability = 0.87 The reliability of the Reading Comprehension Test which was computed through the KR21 Formula, was relatively high which gave assurance to the researchers that they can use the passages for the target group. In order to be able to determine the appropriate readability of the passages for each level, 50 comprehension passages were randomly selected from the TOEFL Test. The readability of each passage was computed through the Fog Index Formula. The mean and the standard deviation of these readabilities were also calculated. Then, the limits for the grade bands for z of -1 and +1 were calculated in order to determine the readability of passages for the elementary, intermediate and advanced groups.

The Nelson-Denny Reading Test

The Nelson-Denny Reading Test, as previously mentioned, comprised 113 multiple choice items. The number of subjects who performed on this test was 162. The highest score which could be obtained by a testee was 156 as the Comprehension Section of the Test, which consisted of 38 items, was double weighed. From the 162 answer sheets that were collected, 20 were discarded since no answer was given to the Reading Comprehension Section. The testees whose scores were within the defined elementary range outnumbered 32. The number of testees in the defined intermediate range reached 82 and the number of those in the defined advanced range reached 30. All the three groups received the six comprehension passages; however, for each group only the two related passages (one intact and the other loaded with deixis)were scored. Since the researchers had to run a two-way ANOVA, it was wiser to have equal sample sizes. Therefore, the number of examinees in the elementary group was adjusted to 30 by eliminating the two middle scores. The number of the examinees in the intermediate group was adjusted to 30 by eliminating the 26 highest and the 26 lowest scores. The result was three groups, each having 30 subjects.

The Reading Comprehension Test

Two sets of data were available for each group. The first set was the examinees' scores on the passages kept intact, and the second set was the examinees' scores on the passages loaded with deixis. The mean and the standard deviation for both sets of data were calculated for each group. The mean and the standard deviation of the first set of data for the elementary group were: Mean (XE1) = 3.93 Standard Deviation (SDE1) = 2.13 The mean and the standard deviation of the second set of data for the elementary group were: Mean (XE2) = 4.73 Standard Deviation (SDE2) = 2.43 For the intermediate group the mean and the standard deviation for both sets of scores were also calculated: Mean (XI1) = 1.73 Standard Deviation (SDI1) = 1.28 Mean (XI2) = 1.63 Standard Deviation (SDI2) = 1.33 The resulting calculated mean and the standard deviation of the two sets of scores for the advanced group were: Mean (XA1) = 2.00 Standard Deviation (SDA1) = 0.91 Mean (XA2) = 1.23 Standard Deviation (SDA2) = 1.10

Results and Interpretations

For the investigation of the first null hypothesis, a two-way ANOVA or a Factorial Analysis of Variance for equal sample sizes was conducted. The data were the scores of the three groups on the two types of the passages (intact and loaded with deixis). The results are in Table 1. See http://rapidintellect.com/AEQweb/sum2004.htm Table 1 indicates that the obtained F for Level appears to equal 54.27 (FL = 54.27). As the critical value of F(2,174) at 0.05 level of significance (=0,05 is equal to 3.05 (F.05(2,174) at 0.05 = 3.05, we notice that the obtained F for Level exceeds the critical value of F at .05 level of significance. Thus, the corresponding null hypothesis is rejected and it is concluded that there is a significant difference among the different levels of subjects in comprehending the two Types of Passages.

Finally, regarding the interaction of the two main variables; that is, source PXL, it is demonstrated in Table 1 that the obtained value of F for PXL is equal to 3.49 (FPXL = 3.49); and the critical value of F(2,174) at 0.05 level of significance (= 0.05) is equal to 3.05 (F.05 (2,174) = 3.05). Thus, the corresponding null hypothesis is rejected and it is concluded that the interaction of the two main variables (Type of Passage and Level) does have a significant effect on the comprehension of the two Types of Passages (intact and loaded) by EFL learners at different levels of language proficiency.

From Table 1 and what was discussed above, we see that Fobs for Passage did not exceed the Fcritical (Fobs = 0.00837 < Fcritical = 3.9), therefore, it is not possible to reject the corresponding null hypothesis and it is concluded that averaged across (ignoring) Level, there was no significant difference between the comprehension of the passages without deixis and those loaded with deixis. However, the observed F value for Level and the interaction of Level and Passage (PXL) did exceed their appropriate critical values (Fobs = 54.27>Fcritical = 3.05) and (Fobs = 3.49>Fcritical =3.05). These results, thus, lead to the rejection of corresponding null hypothesisfor Level and PXL. In other words, for Level, it can be concluded that, averaged across (ignoring) the Type of Passage, reading comprehension varies with the subjects' Level. Regarding the interaction of the two main variables, that is, Level and Passage (PXL), it can be concluded from the above mentioned results that there exists a significant difference in this analysis. In other words, the interaction of the two main variables has a significant effect on the comprehension of the subjects. The best way to interpret an interaction, as Howell (1989) maintains, is to plot the means of one variable separately for each level of the other variable. See http://rapidintellect.com/AEQweb/sum2004.htm

It is quite interesting that for the elementary group, the mean on P2(passage with deixis) is greater than the mean on P1 (intact passage); whereas for the intermediate and advanced levels, the means on P1 are greater than those on P2. However, further investigations are required to find out whether these differences are significant or not. Howell (1989) maintains that whenever the lines are (significantly) non-parallel, there is an interaction and that in many cases the interaction term may well be of greater interestthan the main effects (the effects of variables taken individually).

   A very important technique for analyzing data that contain
   significant interaction is the examionation and testing of
   what are called simple effects. A simple effect is defined
   as the effect of one variable at one level on the other variables.
   (Howell, 1989: 2589)

Benefiting from Howell's explanation, the analysis was adjusted to the case of this research. Therefore, the overall difference between Pl and P2 (passages without and with deixis, respectively) is a main effect, but the difference between P1 and P2 for only those people in the elementary group is a simple effect. Then, the sum of squares and mean square for this simple effect can be computed. To do thesecomputations, we merely pretend that we have data only for people involved in the elementary level. The results of these computations are represented in Table 3. See issue website http://rapidintellect.com/AEQweb/sum2004.htm Table 3 indicates that there is no difference in reading comprehension of passages loaded with deixis and those without deixis for the elementary level. In other words, the difference between means 3.93 and 4.73 is not significant. As a second step, the difference between 1 and P2 can be computed for those people in the intermediate group. The same procedures are to be followed for this simple effect. The esults are represented in Table 4. See http://rapidintellect.com/AEQweb/sum.004.htm

Once again, the value of Fobs does not exceed the Fcritical and the Ho cannot be rejected. The conclusion is that there is no significant difference in reading comprehension of passages loaded with deixis and those without deixis at the intermediate level. In other words, the difference between means 1.73 and 1.63 is not significant. The results of the same analysis for the advanced level are represented in Table 5. See issue website http://rapidintellect.com/AEQweb/sum2004.htm

For the advanced level, also the Fobs does not exceed the F criticaland, therefore, it is concluded that there is no significant difference in reading comprehension of passages with deixis and those without deixis at the advanced level. In other words, the difference between mean 2 and 1.23 is not significant. Now, the same concept of simple effect will be interpreted for Level in order to test the second main null hypothesis (Ho2). Benefiting from Howell's explanation once again, the overall difference among the elementary, intermediate and advanced levels is considered as a main effect, but the difference among these three levels only for the passages loaded with deixis is a simple effect. Then, the sum of squares and the mean square is to be computed for the three groups. The results are indicated in Table 6. See issue website http://rapidintellect.com/AEQweb/sum2004.htm

It is clear from Table 6 that the Fobs exceeds the Fcritical (Fobs = 40.78>Fcritical = 3.5) at .05 level of significance. Therefore, the second null hypothesis is rejected and it is concluded that there is a significant difference among the three levels in comprehending the passages loaded with deixis. However, in order to be able to determine where the difference lies, a comparison should be made between the individual means of the three groups. The procedure for such comparison between individual means is often referred to as the protected t or Fisher's least significant difference test. The first requirement for a protected t is that the overall F for an

Analysis of Variance must be significant and this requirement is met here. The procedure for the computation of the protected t is simply to replace the pooled variance estimate (S2) in the standard t formula by MSerror from the overall Analysis of Variance. In comparing among several groups, we use MSerror instead of S2 because it is based on variability within all the groups rather than within just the two groups we are going to compare (Howell, 1989, p. 237). The results of the protected t-test run between the elementary and the intermediate levels for Type 2 passages are in Table 7. See http://rapidintellect.com/AEQweb/sum2004.htm

The df to be considered here is the dferror which is 174. As it is evident from Table 7, the value of tobtained is greater than tcritical (Tobs(174) = +1.962), and the null hypothesis is rejected at 0.05 level of significance. That is, the elementary group has comprehended the passages loaded with deixis better than the intermediate group. The results of the protected t-test run between the intermediate and the advanced levels for Type 2 passages are in Table 8. See http://rapidintellect.com/AEQweb/sum2004.htm From Table 8, we can see that the value of tobtained does not exceed the tcritical (Tobs.(174) = +1.962), and the null hypothesis is not rejected at 0.05 level of significance. In other words, the intermediate group has comprehended the passages loaded with deixis better than the advanced group.

Conclusion

The purpose of this study was to determine whether there existed any significant difference between the comprehension passages with and without deixis by EFL learners at three different levels. Considering the means of the three groups on P1 and P2, only the elementary group had a greater mean and the intermediate and the advanced groups demonstrated to be higher at PI than P2. However, the conducted statistical analysis on the data highlighted an evidence for the fact that these differences were not significant, and that the comprehension of the two passages were the same within each level. Nonetheless, the statistical analysis conducted for the second null hypothesis was able to reject this hypothesis and provide the evidence for the fact that there existed a difference among the three levels in comprehending the passages loaded with deixis. However, the protected t-test justified that, contrary to the expectations, the elementary group had the best comprehension of Type 2 passages among the three levels. The intermediate group stood at the mid-point and the advanced group had the worst comprehension of the Type 2 passages. Moreover, it should be noted that the difference between the elementary and the intermediate was much greater than the one between the intermediate and the advanced groups. It seems as if the background knowledge of the examinees overcame the hampering effect of the deictic expressions. It is probable that their schemata and world knowledge and the pragmatic knowledge residing beyond the content and the context has disambiguated the lack of clarity of the text created by the decitic expressions.

Pedagogical Implications

In achieving the immediate goals in the EFL/ESL reading classroom, teachers must strive for an optimum balance between the background knowledge presupposed by the text and the background knowledge the students possess. The long-term goal of the reading teachers, nevertheless, must be to develop independent readers outside the EFL classroom, readers whose purpose in learning to read in English as a foreign language is to learn from the texts they read. In order to meet these goals, while deciding about the content of teaching, teachers should be aware of the fact that deixis or deictic terms are variables affecting the EFL readers' comprehension ability. They should also realize that pragmatic residues play an important role in textual comprehension and decitic expressions can paralyze the readers to comprehend the intended meaning of the message.

The goal of reading is to interpret the writer's meaning. This aspect of meaning is not solely derived from the meanings of the words used in phrases and sentences. The readers should try to understand not only what the words mean, but what the writer of those words intend to convey. This emphasis on the theoretical on the role of context should not be left on the theoretical side of the EFL teaching, it should rather be taken into the teaching situation and to the awareness of the students. We have come to the consensus that in the understanding of meaning and in our attempts to communicate, context is a necessary component. Hence, teachers must be aware that not only in teaching oral skills, but also in teaching reading, context and pragmatic parameters should be presented appropriately to the students. The importance of context and its components, many of which are indeed pragmatic residues, must be brought into the consciousness of the students. The process of identifying and dealing with deictic expressions in a passage should make the EFL students more sensitive to such terms and their proper interpretations when they read on their own.

References

Alderson, C. & Urquhart, H. (1984). Reading in a Foreign Language. London: Longman.

Bar-Hillel, Y. (1971). Pragmatics of Natural Languages. Dordrecht: Reidel.

Barnitz, J. (1978). Reading and development of nonnative speakers of English: Research and instruction. Language in Education Series, ERIC Clearinghouse on Language and Linguistics.

Best, J. (1977). Research in education. Englewood Cliff: Prentice-Hall, Inc.

Bouton, L. (1996). Pragmatics and language learning. In Pragmatics and Language Learning. Vol 7, pp. 1-20. Urbana: University of Illinois.

Carrell, P., Devine, J. & Eskey, D. (1988). Interactive approaches to second language learning. Cambridge: Cambridge University Press.

Carnap, R. (1947). Meaning and necessity. Chicago: University of Chicago.

Carton, A. (1976). Orientation to reading. Rowley; Newbury House Publishers, Inc.

Fillmore, C. (1966). Deietic categories in the semantics of "come." Foundations of Language. Vol, pp. 219-227.

Goodman, K. Reading, writing, and written texts. A transactional sociolinguistic view. In Ruddell, R., M. Ruddell and H. Singer. Eds.). (1994). Theoretical models and processes of reading. (fourth ed.). Newark: International Reading Association pp. 1092-1129.

Green, G. (1996). Pragmatics and natural language understanding. (Second ed.). Mahwah; Lawrence Erlbaum Associates, Publishers.

Harman, P. (1990). Teaching indirect speech: deixis points the way. ELT Journal, Vol. 44N3, pp.230-238. Oxford: Oxford University Press.

Hatch, E. & Farhady, H. (1981). Research design and statistics for applied linguistics. Englewood Cliff: Newbury House Publishers.

Howell, C. (1989). Fundamental statistics for the behavioral sciences. Boston; PWS-KENT Publishing Company.

Levinson, S. (1983). Pragmatics. Cambridge: Cambridge University Press.

Morris, C. (1938). Foundations of the theory of signs. In International Encyclopedia of Unified Science. (Ed.). Chicago: University of Chicago

Richards, J., Platt, J. and Webber, H. (1985). Longman dictionary of applied linguistics. Hong Kong: Longman.

Rosenblatt, L. The transactional theory of reading and writing. In Ruddell, R. and M. Ruddell and H. singer. Theoretical models and processes of reading. (Fourth ed.). Newark: International Reading Association, pp. 1057-1091.

Rumelhart, D. Toward an interactive model of reading. Ruddell, R. M. and H. Singer. (Eds.). Theoretical models and Processes of reading. (Fourth ed.). Newark: International Reading Association, pp. 864-893.

Smith, F. (1989). A pragmatic view of French deixis. York Papers in Linguistics. Vol. 14, pp. 760-786.

Stanovich, K. (1980). Toward an interactive-compensatory model of individual differences in the development of reading fluency. Reading Research Quarterly, vol. 16, pp. 32-71.

Taneri, M. (1989). Socially deictic use of body-part term in Turkish. Kansas Papers in Linguistics, vol., 14, PP. 20-157. of Child Language, vol. 10, n1, pp. 23-33.

Varzegar, M. (1993). The componential features of comprehension in reading. Tabriz: Tabriz University Press.

Varzegar, M. (1995). The discoursal aspects of reading comprehension. Pazhuhesh. Vol 11, pp. 109-120.

Varzegar, M. (1997) The Impact of extroversion-introversion on EFL reading comprehension. Tehran: Journal of the College of Foreign Languages.

Varzegar, M. (1998). Authenticity vs. simplification and EFL reading comprehension. Paper presented at Illinois TESOL Convention: Chicago.

Varzegar, M. (2001). Academic competence and teaching reading. Paper Presented at International Reading Associaition Convention. New Orleans.

Varzegar, M. (2003). The features of the compensatory-interactive reading comprehension techniques. NJTESOL/BE. Somerset, NJ.

Wallace, C. (1992). Reading. Oxford: Oxford University Press.

Webber, L. (1991). Structure and ostension in the interpretation of discourse deixis. Language and Cognitive Processes, vol N2,pp. 107-135.

Yule, G. (1985). The study of language. Cambridge: Cambridge University Press.

Yule, G. (1997). Pragmatics. Cambridge: Cambridge University Press.

Minoo Varzegar, Rutgers University, New Jersey

Ali Afkhami, Tehran University, Iran

Mona Khabiri, Tehran University, Iran

Varzegar, Ph.D. from the University of Illinois, published 23 books and 55 articles, Lecturer at the Program of American Language Studies and the English Department. Afkhami, Ph.D. from the University of Paris, Dean of the College of Foreign Languages and Dean of the College of Humanities, Associate Professor. Khabiri, M.A. from Tehran University.