How does the order of introducing the past tense to students affect their learning? Should students be taught the irregular past tense separately from the regular past tense until students have gained a level of mastery? A deeper question raised is why teachers ignore theory that they have embraced in the classroom when it comes to implementation. The data collected by my work sample would suggest that a greater reliance on theoretical research rather than my approach, which was most likely formed due to my bias as a native speaker not recognizing the complexity of acquiring the “simple” past tense, might have resulted in greater student learning.
My work sample focused on teaching an introductory unit over a two-week period on the past tense to eight intermediate and early advanced English as a Second Language (ESL) first graders. The order of presentation was regular past tense, irregular past tense, and past progressive tense. Nearly equal time specifically targeted instruction on the regular past tense and irregular past tense; however, the results on three different types of assessments suggest that the irregular past tense was used and recognized significantly better by students. The regular past tense was used much less frequently than the irregular past tense by students in open-ended situations both in oral and written responses that elicited the past tense and less growth was noted in test questions that required the regular past tense compared to those requiring the irregular past tense on both oral and multiple choice questions.
The rationale for this study has several layers. There is a limited amount of time for instruction and time should be used effectively to promote as much learning as possible. Quality of instruction is important because students will experience political, social, and economic forces that can result in discrimination for those who do not speak Standard English. Students will be perceived as more educated, intelligent, and employable if their language skills are considered excellent. The current curriculum adopted by my district does not provide sufficient practice and instruction to promote mastery and, thus, the teacher needs to be an active agent in providing students with the best instruction, rather than passively following a curriculum. I had attempted to provide a better route for students, and I believe that student success was increased due to my unit of instruction. A real question, though, is: “Could I have done better?
In the reflection on teaching the unit I suggested both an expanded four-week unit and a new sequence for lessons to improve the learning of future students. The irregular past tense would be introduced before the regular past tense based on theories of second language acquisition (SLA) encountered in an introductory linguistics class. This order of SLA was not intuitive for me and perhaps that is why I had subconsciously rejected it. I had determined that it would be easier for my students to learn the regular rule for past tense formation first and then the exceptions. Relying on my own experience as a native speaker may have been doing a disservice to my students.
After looking more closely at the data with more distance, I began to question if I had gone far enough in my revised unit plan. I would like to engage in research to confirm my recent hypothesis that teaching the irregular and regular past tenses separately will result in better student learning or to make a decision to provide additional lessons devoted to practicing the regular past tense beyond what I had originally proposed. Now that I have looked at the data from a different angle, I believe that one of these decisions will significantly and positively affect student learning, rather than my initially proposed revision because at that time I had not recognized that students really had not succeeded in learning the regular past tense.
III. Literature Review/Synthesis
Lightbown and Spada (2006) summarize the research on accuracy order for the development of grammatical morphemes in English. First language (L1) research began with Brown’s (1973) longitudinal study of three children where he identified fourteen grammatical morphemes that occurred among them in a very similar sequence. Many studies beginning in the 1970s have shown that there are differences in acquisition between L1 and second language (L2) learners. However, the order is similar among L2 learners from different L1 backgrounds, though L1 does have an influence on acquisition sequences. For both, L1 and L2 learners, the irregular past forms are acquired before the regular past. Krashen’s (1977) acquisition order suggests four stages of acquisition, though those within each stage may occur in any order,: (1) -ing, plural, and copula, (2) auxiliary and article, (3) irregular past, and (4) regular past -ed, third person singular -s, and possessive ‘s. The use of obligatory contexts (the places in a sentence where the morpheme is necessary to make the sentence grammatically correct) alone rather than also including the incorrect use of morphemes explains some possible discrepancies between the model and real use by L2 speakers. According to Goldschneider and DeKeyser (2001), many studies suggest that children and adults manifest very similar acquisition orders and that the type of instruction has no influence on the order of acquisition of grammatical morphemes, and they do not debate these finding.
Goldschneider and DeKeyser’s (2001) meta-analysis investigated a causal explanation for the “natural” order of acquisition of English grammatical morphemes for L2 speakers. They looked at a combination of five determinants: perceptual salience (how easy it is to notice the morpheme), semantic complexity, morphophonological regularity, syntactic category, and frequency. Their results show that phonological salience and syntactic category have the highest correlation with percentage correct, but that there is a high degree of intercorrelation among the predictors. Many studies have not been able to suggest a single cause for this order, so this new direction in research provides another hypothesis to be tested. As their study suggests, recognizing the causes underlying the order will allow for teachers to increase the rate of acquisition by presenting material or morphemes in a way that capitalizes on these causes.
Lee (1995) challenges the traditional dualistic perspective that regular inflection is learned by rule and irregular inflection by rote. Since irregular verbs occur in families of similar verbs and that even children as young as four can generalize irregular forms to novel verbs, processing is more interactive. Irregular forms can be inflected through: vowel change, final consonant change, vowel and final consonant change, complete change, and no change.
The title “simple” past is misleading and its complexity is explained by both scholars and educators. Rather than just focusing on the grammatical aspect, Haznedar’s (2007) research addresses the four different lexical aspects that the simple past tense addresses: achievement, activity, accomplishment, and state. Riddle (1986) notes that the past tense commonly occurs in both the completive and noncompletive senses and stresses the importance of context, thus making the past tense challenging even for advanced level L2 students. The definition “true before speech time in the real world or in the speaker’s belief world” may be helpful to realize that the actual meaning and discourse functions determine whether present or past tense is appropriate.
Dabaghi and Tavakoli (2009) raise the question why irregular past tense forms are acquired before regular past tense forms despite the fact that the regular rule applies to more verbs than the irregular form, has more reliability, and a higher frequency of input for regular forms. One overlooked fact in other studies is that irregular forms occur very frequently in the input and cite sources that nine out of twelve of the most common verbs are irregular.
IV. Data Findings
The three studies I located that are concerned with the results of students learning irregular and regular past tense were done with university students in Britain and Iran (Lee, 1996; Dabaghi & Tavakoli, 2009) and grades 2-8 in Canada (Herman & Flanigan, 1995). One study was conducted in an English as a Foreign Language (EFL) context with intermediate level students (Dabaghi & Tavakoli, 2009), another with proficient non-native speakers (Lee, 1996), and the third with intermediate to advanced children in an ESL elementary school setting (Herman & Flanigan, 1995). .
Dabaghi and Tavakoli (2009) studied 56 intermediate level students of EFL in Iranian private language school settings. Students read passages at an independent level, engaged in oral reconstruction tasks, and took individualized tailor-made multiple choices based on student errors. Teachers used either immediate explicit, immediate implicit correction, or delayed explicit correction during the oral reconstruction tasks. One of their findings was that when corrective feedback is given, the irregular past tense form is learnt before the regular past tense. So, they suggest that teachers consider the same reasons for better learning of irregular forms in the teaching of regular forms. Teachers can present the regular past tense forms in need of correction as unanalyzed chunks so that they can be learned through memorization, consider that self-correction may be a better alternative for focusing on the correction of regular forms, and make the target feature salient enough during instruction because the more prominent a language form at input, the greater the chance it will be acquired.
In Herman and Flanigan’s (1995) study, 11 students aged 7-14 were given daily formal instruction for two weeks in use of the past tense and plural noun forms in a content-based and communicatively oriented ESL program. A significant difference was found only for the instructed group as far as detection and correction of noun plural forms and not past tense forms when compared to a control group. The students were to read a story silently, locate the forms to be inflected, circle them, and then write the correct forms. Practice was provided for this new test format. Irregular past tense verbs were counted correct only if the entire spelling was correct, whereas for regular past tense only the “-ed” was required to be spelled correctly. There was a ratio of three irregulars wrong for every two incorrect regular past tense forms. Their larger question is whether attention to form is desirable for students at the elementary school level and they concluded that “focus on form” is important.
In Lee’s (1995) study, ten proficient non-native speakers of English and ten native speakers of standard British English were tested for responses to pseudo-irregular and –regular verbs according to their phonological distance to known irregular and regular paradigms. The task was to complete a cloze passage containing a short story for 48 pseudo-verbs and 12 pseudo-adjectives. There was an overall preference for suffixed past tense forms. Results indicated that both irregular and regular forms were produced using analogies by both native and non-native speakers. The rhyme effect was judged to be more crucial than the alliteration effect in triggering analogies. The dynamic interaction of rote, analogy and rule in language learning/processing and cognition is the general model proposed by Lee (1995).
Based on three different types of assessment, my work sample showed that the irregular past tense was used significantly more often by students in nonobligatory contexts. For the open-ended question on the listening-speaking test, “What did you do after school yesterday?,” there were only two uses of the regular past tense, one on the pre-test and two on the post-test compared to the nine uses of the irregular past tense. On the writing test, where students were asked to write about what they did yesterday, only three students out of the seven used the regular past tense and their work included errors. In contrast, all but one student, who is at the early intermediate level as measured by the Oregon English Language Proficiency Assessment (ELPA), used the irregular past tense, five students with errors and two with no errors.
Students improved more in obligatory contexts that required the irregular past tense than those that required the regular past tense. Results were little or moderate improvement on the questions on the listening-speaking test that created an obligatory context to use the irregular past tense in their answers, but no improvement for responses that required the use of the regular past tense. On the multiple choice test, seven students selected “went” appropriately on the post test, showing a gain of three points and all eight students selected “were” appropriately, showing a gain of four students. In contrast, only four students selected “walked,” showing a gain of only one student. An additional factor here might be that students did not recognize “on Sunday” required the past tense, so selected the present tense. It is interesting to note that only one student selected the wrong spelling “walkt” on the post test whereas three had done so on the pre test, which may suggest that some students did learn that “walkt” is not a real word. However, the question that was designed to solicit the response “played” did not present any such ambiguous context. Six students selected the correct answer on the post test, showing only a moderate gain of two students.
Scores on the ELPA for the intermediate and early advanced students in the group did not seem to be helpful predictors of student use of the regular and irregular past tense. It did seem to accurately measure the expected level of knowledge of the one early intermediate student. This student was able to use only one past tense verb on the listening-speaking portion, “went,” and none on the written test. However, this student did gain three points on the reading-writing test, suggesting that instruction made some difference in learning. Overall this student did not appear ready to acquire the simple past tense, as the ELPA score would indicate.
V. Data Analysis
Due to the previously discussed differences between the three studies on regular and irregular past tense forms and my context, the studies cannot be compared directly with my work sample. I find reason to question some of the results of Herman and Flanigan’s (1995) study, which seem to challenge previous SLA research. Since they required the correct spelling of irregular past tense verbs, this may have skewed their results. Is proper spelling considered to be the true measure of knowledge of irregular forms? I did not find that documented in other studies. On the other hand, two weeks of instruction time, did not appear to result in significant learning, which does correspond with my results.
Two studies required obligatory context where students had to correct errors in a text (Herman & Flanigan, 1995), or complete a cloze passage with pseudo-verbs (Lee, 1995). Dabaghi and Tavakoli’s (2009) study focused on oral re-telling which allowed for more flexibility in student response. My work sample had an obligatory context for the past tense in all assessment questions, but only in selected multiple-choice and oral questions was the irregular or regular form obligatory. The overall design of my assessment may be helpful in the future to determine what students have acquired due to the comprehensive nature of the assessment. Assessments that target the past tense would be important, in most cases, to determine specific instruction and grouping students for instruction, rather than relying on the students’ ELPA scores or simply following the curriculum guide.
In teaching this unit in the future with first grade ESL students, I would consider several things that would require prior assessment of students’ use of grammatical morphemes. First, have students already learned grammatical features that occur higher on the SLA scale? If not, they may not be ready for targeted past-tense instruction, as there has been no research that has strongly debunked the acquisition sequence developed in the 1970s. Second, what forms of the past tense, if any, do the students use? If students are not yet using the past progressive tense, that would be the starting place. If students are not using the simple past tense, then I would restrict the unit to the irregular tense and expand it. If students know several irregular past tense verbs, then I would begin with irregular past tense and move to regular past tense.
In irregular past tense instruction, I would take advantage of rhyming to help students learn more irregular verbs. Verbs would be grouped during instruction to facilitate learning patterns in irregular verbs, such as the eight common irregular verbs that rhyme with “bought” (though spelling patterns differ). My focus would be student learning of the regular past tense in chunks first. I would allow students, when they are ready, to analyze how the endings are formed rather than present a rule. This better fits with my constructiveness approach to education.
Celce-Marcia (1991) proposes a grid to help the teacher determine the degree to which it is appropriate to focus on form with a given group of students, which I will find helpful. The six variables are: age, proficiency level, educational background, skill, register, and need/use.
The results of my research report can be presented to ESL teachers in my district to challenge the curriculum, especially since the past tense is a significant element of ESL instruction at the elementary school level. There is wide agreement that students acquire the irregular past tense first, so instruction time is wasted when the acquisition sequence is not considered and the regular past tense is presented in the curriculum before the irregular. Attention can be given to grouping students for instruction based on readiness to acquire grammatical morphemes. The complexities of the past tense discovered during this research will assist in fleshing out what is required to learn/acquire the past tense. Some suggestions for making instruction of both the regular and irregular past tense more useful are also gained through this research. Looking to multiple causes for acquisition of grammatical morphemes (Goldschneider and DeKeyser, 2001) and the interaction of the methods of rote, rule, and analogy with grammatical morpheme acquisition (Lee, 1995) will help teachers to provide instruction for all students to reach mastery in using the past tense.
Celce-Murcia, M. (1991). Grammar pedagogy in second and foreign language teaching.
TESOL Quarterly, 25(3), 459-480.
Dabaghi, A. & Tavakoli, M. (2009). A comparison of the effects of corrections on
definite/indefinite articles and regular/irregular past tense forms: A case of Iranian
EFL learners. Asian EFL Journal, 11(4), 90-114.
Goldschneider, J. M. & DeKeyser, R. M. (2001). Explaining the “natural order of L2
morpheme acquisition” in English: A meta-analysis of multiple determinants. Language
Learning, 51(1), 1-50.
Haznedar, B. (2007). The acquisition of tense-aspect in child second language English. Second
Language Research, 23(4), 383-417.
Herman, R. L. & Flanigan, B. O. (1995). Adding grammar in a communicatively based ESL
program for children: Theory in practice. TESL Canada Journal, 13(1), 1-16.
Lee, B. (1996). On the processing of regular and irregular inflections: The symbolist-
connectionist debate revisited. In: Koster, C. & Wijnen, F. (Eds.), Proceedings of the
Groningen Assembly on Language Acquisitionheld at the University of Groningen, 7-9
September 1995, (pp. 273-282). Groningen, The Netherlands: The Centre for Language
Lightbown, P. M., & Spada, N. (2006). How languages are learned (3rd edition).Oxford:
Oxford University Press.
Riddle, E. (1986). The meaning and discourse function of the past tense in English. TESOL