Preterite had +V- ed : A Developmental Narrative Structure of African American English

In this study, we examined the use of had + V-ed in the language samples of four- and six-year-old children using the framework of Rickford and Rafal (1996). Specifically, we asked whether children who speak African American English (AAE) used had + V-ed to refer to simple past tense within narratives. Based on language samples from 93 children (40 speakers of AAE and 53 speakers of a rural version of Southern White English [SWE]), our results were consistent with Rickford and Rafal's claims. About half of the AAE speakers (and none of the SWE speakers) produced had + V-ed as a preterite, and these forms frequently occurred in the complicating action clauses of narratives. The AAE speakers' use of preterite had + V-ed also increased with age and was directly related to narrative skill.

Among the highly proficient child narrators in this study, the six-year-old who produced the greatest number of complex narrative sequences also provided the most tokens of preterite had + V-ed. One of her narratives, containing 4 occurrences of preterite had + V-ed, appears in (1).

1. a. My mama, she was about to go to Bible study,
b. and on the way back there, her car HAD STOPPED.
C. And then she HAD CALLED the house because somebody let her use the phone.
d. And then she HAD CALLED the house,
e. and then I said, "Hello. Who's this?"
f. And then my mama said, "It's your mama. Let me talk to your daddy."
g. And then she HAD TOLD my daddy to come with us and bring a big rope so they could pull the car home.
h. So, we got a new car.

In 1996 Rickford and Rafal presented 52 occurrences of had + V-ed that were used in an unexpected preterite sense by nine African American preadolescents, aged 11-13 years, from East Palo Alto, California. In standard American English, had + V-ed combinations reflect the pluperfect, as in (2).

2. I HAD BOUGHT some jambalaya by the time the crawfish came.

As discussed by Rickford and Rafal, the pluperfect form had bought establishes that one remote completed event (e.g., buying jambalaya) occurred before another completed event (e.g., arrival of crawfish), and both of these events took place before the utterance itself was produced. The use of the pluperfect is thus associated with three distinct events, each ordered temporally in relation to the others. For this reason, Comrie (1985, 78) labeled the pluperfect as an "absolute-relative" tense. What is absolute is that the pluperfect refers to the past; what is relative is that the past event is prior in time to another past event. In contrast, the preterite tense marks a completed event as occurring prior to the utterance time but without further temporal reference, as in (3).

3. I BOUGHT some jambalaya.

Since it does not orient the completed event time in relation to any other past event, the preterite is an absolute indicator of past time. What distinguishes the two tenses, then, is whether the completed event is relatively more remote than another event.

Within AAE varieties, Rickford and Rafal (1996) argued that the had + V-ed structure reflects preterite tense because in 52 of 54 (96%) occurrences it referred to simple, absolute past rather than remote past. They also demonstrated with their data that preterite had + V-ed does not serve as a compensatory mechanism for zero inflectional marking of preterites, but instead serves a narrative discourse function. Indeed, 100% of their preterite had + V-ed tokens were produced within a narrative, and 94% of these also occurred within the complicating action clauses of the narratives rather than in the narrative abstract, orientation, embedded orientation, result, or coda.

Characteristics of the Language Samples Studied Here

The data for...