Abstract
This article describes a study that examined the development of the past tense counterfactual (PTCF), a complex syntactic pattern, in younger and older adolescents. An example of such a sentence is “If Michael had done all the math homework, he might have passed the final test.” Past tense counterfactual sentences express subtle meanings and have a hypothetical, conditional quality. They imply that certain events did not occur (i.e., Michael did not do all the math homework; Michael did not pass the final math test), but at the same time, they expect the listener or reader to consider the possibility that they could have occurred under different circumstances. Past tense counterfactuals are used to explain, predict, or comment on events, express opinions, or pass judgment on others, as in “If Amy had gone to bed earlier, she would not have fallen asleep in class.” To convey the intended meaning, the speaker or writer must follow specific grammatical rules. The subordinate clause in a PTCF sentence must employ the correct form of the past perfect verb (e.g., had done, had gone), and the main clause must employ a modal verb (e.g., might, would) and the correct form of the present perfect verb (have passed, have fallen).
The PTCF is a complicated sentence for several reasons:
In the subordinate clause, one must select the correct form of the past perfect (e.g., had washed, had published) and then, in the main clause, one must select an appropriate modal auxiliary verb (e.g., might, would), along with the correct form of the present perfect (e.g., have died, have learned) to achieve parallel structure (Crutchley, 2004).
The sentence clauses must be precisely coordinated. The decision to use the present perfect or the past perfect is dependent on the relative timing of past events. The present perfect refers to an event that was completed (or could have been completed) before the present time, the past perfect refers to an event that was completed prior to another past event (Crews, 1977).
Past tense counterfactual sentences have a low frequency of occurrence in spoken language. Consequently, children have limited exposure to this type of sentence, making it difficult for them to learn the correct grammatical patterns and to retrieve them during spoken or written language (Crutchley, 2013; Diessel, 2004).
The authors argue that use of these syntactic structures is an important aspect of academic language, particularly in science, math, and social studies. For example, “If the Navajo code talkers had not developed an unbreakable code, the U.S. might have lost the 2nd World War in the Pacific.” Or “If Edward Jenner had not discovered that people infected with cowpox were immune to smallpox, he would not have invented the first vaccine to prevent illness.”
Given the complexity of PTCF sentences—with their grammatical, temporal, and experiential demands—accurate production of these sentences is a late linguistic attainment that could challenge many adolescents. The authors of this article explored this hypothesis with adolescents by asking them to complete a set of PTCF sentences when provided with just the subordinate clause, thereby examining their ability to produce the present-perfect verb form in the main clause.
Participants
The main participants were two groups of adolescents who were students in 8th grade (26 females, 14 males) and 11th grade (34 females, 6 males) attending racially/ethnically diverse low-income schools. The eighth-grade group had a mean age of 13.20 years and the 11th-grade group had a mean age of 16.58 years. All adolescents spoke American English; however, 18% of the 13-year-olds and 20% of the 16-year-olds reported themselves to be bilingual. All students were in general education classes. No standardized language scores were available, but the researchers did have written language samples for the students from which they were able to obtain the mean length of communication unit (MLCU), an index of general language competence and syntactic development. A control group of 40 young adults were also recruited to establish an adult level of performance on the past tense counterfactual (PTCF) sentences task. This group (37 females, 3 males) had a mean age of 22.88 years, and all were undergraduate or post-baccalaureate students enrolled at a nearby university. For each adult, the researchers had written language samples from which they obtained MLCU scores.
Procedures
All participants were administered a written language task that involved the interpretation of fables in addition to the production of PTCF sentences. The examiner read the practice fable, The House Mouse and the Field Mouse, its moral message, the summary statement, and the incomplete PTCF sentence aloud to the whole class as the participants followed along in their test booklets.
Moral: “The grass is not always greener on the other side of the fence.”
Summary statement was “The field mouse was afraid of the cat and the people who lived in the house.”
Incomplete PTCF sentence: “If the field mouse had not gone home to live in peace . . .” (with the past perfect verb form in bold).
Participants were asked to complete the sentence by writing a sentence that made sense with the story. After the participants wrote a response, the examiner stated that “A good answer would be, ‘He would have been very unhappy,’” and repeated the whole sentence (“If the field mouse had not gone home to live in peace, he would have been very unhappy”), modeling a fully correct PTCF sentence.
The participants then read four more stories and completed sentence stems. To earn full credit for completing a sentence, the response had to use all three elements of the present-perfect verb form correctly: (1) an appropriate modal verb (e.g., would, could, might); (2) the auxiliary verb have; and (3) an appropriate past participle verb. Each element was worth 1 point. The following are example responses and scores for the incomplete PTCF for The Fox and the Grapes story.
He wouldn’t have been mad at the grapes (age 13, 3 points).
He may have been able to fulfill his desire (age 16, 3 points).
He would have delicious grapes (age 13, 2 points; lacks past participle verb).
The fox would have got the grapes and had a great day (age 13, 2 points; lacks past participle verb).
He could reach the grapes (age 13, 1 point; lacks have and past participle verb).
He would be able to quench his thirst (age 13, 1 point; lacks have and past participle verb).
Results
The mean MLCU scores for each group were within the normal range, but there was wide variability among the participants. The young adults had significantly higher MLCU scores than the adolescents, but there was no difference between the two adolescent groups.
On the PTCF, the young adults performed significantly better than the adolescents. Although 16 year olds obtained higher PTCF scores than 13 year olds, the difference was not significant.
The researchers defined mastery as accuracy of 90% or higher (11 of the possible 12 points). Forty-eight percent of 13-year-olds, 60% of 16-year olds, and 73% of 22-year-olds demonstrated mastery.
The three elements of the counterfactuals were mastered at different ages: for the 13-, 16-, and 22-year-olds, respectively, 100%, 95%, and 100% earned full credit for the modal verb; 53%, 63%, and 73% did so for the auxiliary verb; and 38%, 45%, and 55% did so for the past participle.
Past tense counterfactual scores were not correlated with MLCU scores, indicating that PTCF sentences were not associated with general language or syntactic development.
This study demonstrates that correctly producing the grammar of PTCF sentences is a late linguistic attainment, extending even into adulthood. All groups were successful using the modal verb would. The auxiliary verb have and the past participle were more difficult for all participants. However, using the auxiliary verb have in conjunction with an appropriate past participle verb was difficult, especially for the group of 13-year-olds, who often omitted both of those elements. Instead, they often used a modal verb followed by an infinitive, as in could reach instead of could have reached, would be instead of would have been, would have instead of would have eaten, would die instead of would have died, and would get instead of would have gotten. Even when participants used the auxiliary verb have correctly, they often used an incorrect form of the past participle, substituting an irregular past tense verb for the past participle. Thus, they produced responses such as have got instead of have gotten, have ate instead of have eaten, and have flew instead of have flown. All groups had the most difficulty with the past participle.
Incorrect responses may have been due to participants’ difficulty in coordinating the relative timing of past events. With a PTCF sentence, both events happened in the past; however, one event preceded the other, necessitating a back shifting of tense. For this reason, in a grammatically correct PTCF sentence the past perfect verb form in the subordinate clause (e.g., “If the rooster had listened to the wily fox”) requires the present-perfect verb form in the main clause (e.g., “He would have been tricked by the fox”), based on rules of formal English grammar. However, some participants completed the sentences in a way that disregarded this timing issue, instead producing sentences that expressed a conditional future tense in the main clause (“If the rooster had listened to the wily fox, he would get hurt” instead of “If the rooster had listened to the wily fox, he would have gotten hurt”). They were not attending to the timing issue and failing to coordinate the two halves of the sentence to achieve parallel structure.
The authors suggest that precision in PTCF sentences may not be essential in conversational language (as long as you get your point across), but precision becomes important in expository language, particularly in science. Acquiring specialized knowledge and having a need to explain it to others during an oral or written report could require more complex structures such as, “If giant sequoia trees had not developed wide-spreading root systems, they would not have survived for thousands of years.” A student who can employ PTCF sentences would be able to convey complex scientific information with greater precision than a classmate who cannot.