A Discourse Analysis of First CorinthiansRalph Bruce Terry
Previous SectionNext SectionHome | Chapters: 1 | 2 | 3 | 4 | 5 | 6

Participant Reference

To date, most of the work on participant reference has been done on narrative texttype. Longacre (1989b, 141-157) devotes chapter 6 of his study on the Joseph narrative in Genesis to participant reference. Callow (1974, 32-37) discusses participants as they relate to cohesion mainly in terms of the gospel narratives. And Grimes (1972, 43-50) discusses participants in a number of non-biblical narratives. But theme-oriented texttypes (i.e., expository, hortatory, and persuasive) differ from sequential event texttypes (narrative and procedural) in that the themeline is more prominent in the former than in the latter.

To be sure, epistolary texts do contain a storyline as well as a themeline. But the storyline in letters has to do with such items as what the writer has been doing, what the writer hears the reader has been doing, letters that the writer has written the reader, letters that the writer has received from the reader, and plans that the writer has for the future, especially any travel plans that relate to seeing the reader. All of these elements are present in the letter of I Corinthians. But an epistolary storyline has few restrictions on temporal ordering such as a narrative has. And while letters may be written that are primarily epistolary storyline, letters of advice such as I Corinthians have a primary themeline and only a secondary storyline.

This primary themeline has a significant effect on participant reference. First, the central participants are the writer and the readers. This is signaled by use of the first person and second person grammatical categories, respectively. Second, if the letter is primarily concerned with advice, the majority of subject slots refer not to participants or props, but to key concepts within themes. Finally, unless the topic is fairly unified, the themes (as contained in the subject slots) may change quite often so that there are few long chains of reference to one concept. All of these are characteristic of I Corinthians.

Greek has several grammatical devices to point to conceptual reference, whether of participants, props, or themes. These include clauses embedded within subject and object slots, noun phrases, nouns (both with an article and without), vocative nouns, participles, pronouns, articles alone, and verb suffixes. In addition, it is also possible to have null reference, where a grammatical trace is missing, but the reference is obviously to a concept which has already been introduced. This is the regular case with the subject of infinitives, which are often missing but are frequently the same as the subject of the main verb in the sentence. With other verb forms, the verb ending provides a minimal trace to help provide reference to the concept in focus. This being the case, other examples of null reference all are found in object slots. Greek will sometimes omit objects which are obligatory in English, but the co-occurrence possibilities of the available concepts with the verb being used serves to limit the reference. The reader knows what concept is being acted upon in the clause because no other concept makes good sense.

Most of these grammatical devices may be used to switch the reference to a new concept, resume reference to a concept already introduced (whether currently under discussion or not), or continue reference to the same concept that has been the primary focus. But there are certain preferred forms for each of these tasks. Table 21 presents the relationships between the subject slot grammatical devices and these tasks.

TABLE 21

RELATIONSHIP OF GRAMMATICAL DEVICES AND TRACKING TASKS

FOR I CORINTHIANS

                                                                    
|Grammatical | Continuation   | Resumption from | New or Different |
|Device      | of Same Subj.  | Clause or Colon | Reference        |
|clause      |  3.7% (n=20)   |  1.3% (n=2)     |  3.2% (n=22)     |
|            |                |                 |                  |
|noun or NP  |  6.3% (n=34)   | 25.9% (n=41)    | 41.4% (n=281)    |
|            |                |                 |                  |
|vocative    |   .2% (n=1)    |   .6% (n=1)     |   .9% (n=6)      |
|            |                |                 |                  |
|participle  |   .6% (n=3)    |   .6% (n=1)     |  1.0% (n=7)      |
|            |                |                 |                  |
|pronoun     | 12.4% (n=67)   | 20.3% (n=32)    | 19.7% (n=134)    |
|            |                |                 |                  |
|article     |  2.6% (n=14)   |  1.3% (n=2)     |  5.7% (n=39)     |
|            |                |                 |                  |
|verb suffix | 74.2% (n=400)  | 50.0% (n=79)    | 28.0% (n=190)    |
|Totals      |          539   |          158    |          679     |

These tracking tasks can be defined as follows. The first tracking task is the continuation of the same subject as that used in the previous clause. The second task is the immediate resumption of a subject previously mentioned in a previous clause within the same colon or the immediately preceding colon following the introduction of a different subject. The final task is the introduction of a different subject. This category includes both resumption of subjects which have not been used in the current or preceding colons and subjects which are entirely new.

From Table 21 it is possible to see that the three major subject slot devices used to express reference are the noun, the pronoun, and the verb suffix. The frequency of each of these varies markedly depending upon the reference task being carried out. For example, in 74.2% of the cases where the same concept must be referred to, the grammatical device used is the verb suffix. Nouns or noun phrases are used only 6.3% of the time to carry out this task. Pronouns are used in this situation only 12.4% of the time. Where the task is to resume a reference to a concept that has been mentioned in a previous clause in the colon or in the previous colon, pronoun usage increases to 20.3% and noun usage to 25.9%. The usage of 50.0% verb endings to express this resumption does not produce as much ambiguity as might be thought in a language which marks number and person on finite verbs and number and gender on participles. Finally the introduction of a new concept or resumption of a concept not recently referred to is done with nouns in 41.4% of the time. Pronoun usage stays about the same (19.7%), but reliance on verb ending usage alone drops to 28.0%. The chi-square test shows the skewed distribution of grammatical devices and reference tasks to be highly significant.

If the 21 new concepts introduced in subject slots are split out of the last column of Table 21 and looked at alone, 19 (86.4%) are introduced by nouns or noun phrases. One is introduced by a participial clause ("those baptized on account of [ὑπέρ] the dead" in 15:29) and one by the indefinite pronoun τινα 'someone' (in 5:1). No new concepts are referred to only by verb endings. The primary way in which new concepts are introduced in I Corinthians is to position them in object slots (either as direct objects, indirect objects, or objects of prepositions). It is notable that only 21 new concepts are introduced as subjects.

Several other variables are highly significant when compared with the same or different subject. Table 22 shows how the tracking tasks relate to the three major divisions of a colon or sentence. There is no significant difference as to how the tasks are distributed between preceding dependent clauses and independent clauses, but there is a highly significant difference between the distribution of tracking tasks in these types of clauses and in dependent clauses that follow the independent clause. These following dependent clauses show a much higher frequency (47.1% as compared with 37.5% to 38.3%) of retaining the same subject as previously mentioned. In the same way, different subjects are much more likely to be found in preceding dependent clauses (48.1%) and independent clauses (52.3%) than in following dependent clauses (37.7%).

TABLE 22

RELATIONSHIP OF TRACKING TASKS AND SENTENCE LOCATION

FOR I CORINTHIANS

                                                                    
| Tracking Task | Preceding Dep. | Independent   |  Following Dep. |   
| Same Subject  | 38.3% (n=70)   | 37.5% (n=364) | 47.1% (n=105)   |
|               |                |               |                 |
| Resumed       | 13.6% (n=25)   | 10.2% (n=99)  | 15.2% (n=34)    |
|               |                |               |                 |
| Different     | 48.1% (n=88)   | 52.3% (n=507) | 37.7% (n=84)    |

There is a similar situation when the first clause in a colon is compared with the clauses that follow it, as shown in Table 23. Once again, following clauses show a much higher rate (48.4% as compared to 25.3%) of retaining the same subject than first clauses in a colon do. First clauses within colons introduce different subjects in 64.9% of the cases, while following clauses introduce different subjects in only 39.0% of the cases. This distribution of tracking tasks is also statistically highly significant. Since first clauses can occur as either a preceding dependent clause or in an independent one, but not as a following dependent clause, it may be that the distribution shown in Table 23 is responsible for the difference in distribution shown in Table 22 between the following dependent clause and the other two types.

TABLE 23

RELATIONSHIP OF TRACKING TASKS AND CLAUSE LOCATION

FOR I CORINTHIANS

                                                       
| Tracking Task |  First Clause  |  Following Clauses  | 
| Same Subject  |  25.3% (n=140) |  48.4% (n=399)      |
|               |                |                     |
| Resumed       |   9.8% (n=54)  |  12.6% (n=104)      |
|               |                |                     |   
| Different     |  64.9% (n=358) |  39.0% (n=321)      |

There is also a highly significant difference in tracking tasks when they are distributed across different types of colons or sentences (either statement, question, or command), as shown in Table 24. Statements are more likely (41.0% as compared to 34.0% or 36.4%) to continue the same subject as previously than either questions or commands, respectively. Questions are more likely (58.0% as compared with 48.7% or 44.9%) to begin a new or different subject than either statements or commands, respectively. And commands are more likely (18.6% as compared to 10.3% or 8.0%) to resume a subject which was discussed in a previous clause or the previous colon than either statements or questions, respectively.

TABLE 24

RELATIONSHIP OF TRACKING TASKS AND STATEMENT OR QUESTION

FOR I CORINTHIANS

                                                                    
|  Tracking Task  |  Statement     |  Question     |  Command      | 
|  Same Subject   |  41.0% (n=381) |  34.0% (n=68) |  36.4% (n=90) |
|                 |                |               |               |
|  Resumed        |  10.3% (n=96)  |   8.0% (n=16) |  18.6% (n=46) |
|                 |                |               |               |   
|  Different      |  48.7% (n=452) |  58.0% (n=116)|  44.9% (n=111)|

Texttype is another variable which shows a significant difference in the way that tracking tasks are distributed throughout a text. Table 25 shows that the narrative texttype embedded in I Corinthians is significantly different from the non-narrative texttypes. It shows a 54% rate of retaining the same subject, while the most that any non-narrative texttype shows is a 40% retention rate. Conversely, clauses in narrative texttype begin new or different subjects only 33% of the time, while non-narrative texttypes show rates of 47% to 57% for this tracking task. But the significance of Table 25 is not limited to the difference between narrative and non-narrative texttypes. Even the differences between the hortatory and persuasive texttypes (which seem to be the most similar) are statistically significant. And expository texttype shows markedly less (33% as compared with 39% or 40%) of the same subject tracking and markedly more (57% as compared with 47% or 51%) of different subject tracking than either hortatory or persuasive texttypes.

TABLE 25

RELATIONSHIP OF TRACKING TASKS AND TEXTTYPE

FOR I CORINTHIANS

                                                                           
|  Tracking Task  |  Narrative | Hortatory   |  Persuasive |  Expository  | 
|  Same Subject   | 54% (n=31) | 39% (n=261) | 40% (n=185) | 33% (n=62)   | 
|                 |            |             |             |              |     
|  Resumed        | 12% (n=7)  | 14% (n=92)  |  9% (n=40)  | 10% (n=19)   |
|                 |            |             |             |              |     
|  Different      | 33% (n=19) | 47% (n=317) | 51% (n=238) | 57% (n=105)  |

Some of this difference in texttypes as far as tracking tasks is concerned may be due to the difference in tracking task distribution as it relates to subject person. Table 26 shows how tracking tasks relate to first, second, and third person subjects, respectively. Note that third person subjects (which characterize expository texttype) show a highly significant difference from either first or second person subjects when compared on either the same or different subject tasks. First person (with 58.3%) has more continuance of the same subject than second person (41.7%), which has more continuance of the same subject than third person (31.2%). The use of the first person (with 28.7%) is much less likely to introduce a new or different subject than either second person (with 42.1%) or especially third person (with 59.1%).

TABLE 26

RELATIONSHIP OF TRACKING TASKS AND SUBJECT PERSON

FOR I CORINTHIANS

                                                                       
|  Tracking Task  |  First Person  |  Second Person  |  Third Person  | 
|                 |                |                 |                |
|  Same Subject   |  58.3% (n=183) |  41.7% (n=90)   |  31.2% (n=261) |
|                 |                |                 |                |   
|  Resumed        |  13.0% (n=41)  |  16.2% (n=35)   |   9.7% (n=81)  |
|                 |                |                 |                |   
|  Different      |  28.7% (n=90)  |  42.1% (n=91)   |  59.1% (n=494) |

Perhaps this difference accounts for the fact that of the fifty-seven chains of four or more consecutive subject references to the same concept in I Corinthians, twenty-two use the first person and fourteen use the second person. These chains are listed in Table 27. This table shows both the length of some reference chains and the grammatical devices used to maintain those chains.

The first column in Table 27 lists the verses in which the chains begin. The size column gives the number of clauses for each chain. The next column lists the tracking task of the first clause to refer to the subject of the chain. There are four possible beginning types: new, different, colon, and clause. New is the first introduction of a topic to the letter. Different is the use of a topic which has not recently been used as a subject. Colon is resumption of a topic which appeared in a subject slot in the previous colon. And clause is resumption of a topic which appeared in the subject of a previous clause within the same colon. The fourth column gives the person and number of the subject being tracked. Where the person and number is changed within the chain, the change is given after a slash. The next column gives the number of colons over which the chain extends. The final column gives a representation of the grammatical categories that make up the chain. Bold print is used to identify the point or points at which the person and number begin or switch.

The longest same reference chains in I Corinthians are found in either the first person or third person. Four chains of the same reference for fifteen or more subject slots occur in I Corinthians. It is perhaps significant that two of these four chains occur in the peak section of chapters 12 through 15. That is 50% of the very long chains in 25% of the letter. It is not the presence of such chains that is significant, but their frequency.

The first long chain is found beginning in 4:10 and extending for eighteen subject slots over five different colons. It begins with a first person plural pronoun (ἡμει̑ς 'we') and extends for fourteen more subject slots using first person verb endings only. Then it narrows the focus by shifting to first person singular verb endings for three more subject slots.

TABLE 27

CHAINS OF FOUR OR MORE CONSECUTIVE REFERENCES

TO THE SAME CONCEPT

                                                                            
| START | SIZE | BEGIN.TYPE | PR&NM | COLONS | GRAMMATICAL CATEGORIES       
| 1:12  |  5   |   diff.    | 2s/1s |   1    | ppppp
| 1:26  |  4   |   diff.    | 2p/3p |   1    | vnnn 
| 1:27  |  5   |   diff.    | 3s    |   1    | nenen
| 1:30  |  5   |   diff.    | 1s    |   2    | peeee
| 3:18  |  4   |   diff.    | 3s    |   1    | peee 
| 4:3   |  4   |   diff.    | 1s    |   3    | eeee 
| 4:7   |  9   |   diff.    | 2s/2p |   6    | eeeeeeeee
| 4:10  | 18   |   colon    | 1p/1s |   5    | peeeeeeeeeeeeeeeee
| 6:9   |  5   |   diff.    | 2p    |   5    | eeepe
| 6:11  |  4   |   diff.    | 2p    |   2    | peee 
| 6:19  |  5   |   clause   | 2p    |   4    | eeeee
| 7:10  |  4   |   diff.    | 3s    |   1    | neee 
| 7:25  |  4   |   diff.    | 1s/4s |   1    | eece 
| 7:27  |  4   |   diff.    | 2s    |   3    | eeee 
| 7:36  |  4   |   clause   | 3s/3p |   3    | peee 
| 7:37  |  6   |   diff.    | 3s    |   1    | peeeec
| 7:39  |  5   |   colon    | 3s    |   2    | eeeee
| 8:8   |  4   |   diff.    | 1p    |   2    | eeee 
| 8:13  |  5   |   diff.    | 1s    |   4    | eeeee
| 9:4   |  5   |   diff.    | 1p    |   2    | eeeee
| 9:12  |  4   |   colon    | 1p    |   2    | eeee 
| 9:16  |  5   |   diff.    | 1s    |   3    | eeeee
| 9:18  | 20   |   diff.    | 1s    |   8    | eeeeeeeeepeeeeeeeeee
| 9:26  |  8   |   diff.    | 1s    |   3    | peeeepee
| 10:1  |  6   |   diff.    | 3p    |   2    | nppppe
| 10:7  |  5   |   diff.    | 3s/3p |   1    | neeee
| 10:12 |  4   |   diff.    | 3s    |   1    | aece 
| 10:23 |  4   |   diff.    | 3s    |   2    | pppp 
| 10:31 |  4   |   diff.    | 2p    |   1    | eeee 
| 11:5  |  5   |   diff.    | 3s    |   3    | neene
| 11:22 |  4   |   colon    | 2p    |   2    | eeee 
| 11:22 |  5   |   diff.    | 1s    |   4    | eeepe
| 11:23 |  5   |   diff.    | 3s    |   1    | neeee
| 11:26 |  5   |   diff.    | 2p    |   2    | eeeee
| 11:28 |  6   |   diff.    | 3s    |   2    | neeace
| 11:31 |  5   |   diff.    | 1p    |   2    | eeeee
| 12:1  |  5   |   diff.    | 2p    |   2    | peeee
| 12:15 |  4   |   new      | 3s/1s |   1    | neee 
| 12:16 |  4   |   new      | 3s/1s |   1    | neee 
| 12:29 |  7   |   diff.    | 3p    |   7    | ppppppp
| 12:31 | 15   |   diff.    | 1s    |   4    | eeeeeeeeeeeeeee
| 13:4  | 16   |   diff.    | 3s    |   3    | nneneeeeeeeeeeen
| 13:11 |  6   |   diff.    | 1s    |   2    | eeeeee
| 14:6  |  4   |   diff.    | 1s    |   1    | eeee 
| 14:15 |  4   |   diff.    | 1s    |   2    | eeee 
| 14:18 |  5   |   diff.    | 1s    |   2    | eeeee
| 14:26 |  6   |   diff.    | 2p/3s |   1    | epeeee
| 15:3  |  6   |   diff.    | 3s    |   2    | neeeee
| 15:7  |  4   |   colon    | 3s    |   2    | eeee 
| 15:9  |  6   |   diff.    | 1s    |   2    | ppeeee
| 15:30 |  4   |   diff.    | 1p/1s |   3    | peee 
| 15:36 |  4   |   diff.    | 3s    |   2    | ceee 
| 15:42 |  7   |   diff.    | 3s    |   4    | eeeeeen
| 16:1  |  4   |   colon    | 2p/3s |   2    | ppee 
| 16:5  |  4   |   colon    | 1s    |   2    | eeee 
| 16:6  |  5   |   clause   | 1s    |   2    | eeeee
| 16:13 |  4   |   diff.    | 2p    |   1    | eeee                      
Size is the number of subjects.  Bold type in the last column indicates
a beginning point or shift in person and number.
Key: Person & Number: s=singular; p=plural
Grammatical Categories: n=noun; v=vocative; c=clause
p=pronoun; a=article; e=verb ending

The second long chain of same subject reference begins in 9:18 and extends across eight colons filling twenty subject slots. This is the longest such chain in the letter. With the exception of the tenth subject slot (filled by the pronoun αὐτός 'self', in context meaning 'myself'), all same subject tracking is carried forward by first person singular verb endings.

The last two long chains occur in the peak sections. The third such chain begins in 12:31 and extends across four colons filling fifteen subject slots. It is carried forward entirely by first person singular verb endings. The fourth long subject chain begins immediately after that in 13:4 and fills sixteen subject slots across three colons. It differs from the others in that it carries the third person concept ἀγάπη 'love'. The concept of love has already been mentioned in the direct object slot of the three preceding colons. The chain begins with the noun in the first two subject slots, then a verb ending in the next clause, and then the noun in the fourth subject slot. It continues on with just third person singular verb endings, and concludes in 13:8 with the repeat of the noun in the sixteenth and final subject slot in the chain. It becomes obvious from this that long third person subject chains are more complex than first person subject chains.

From Table 27 it is possible to see that many of the chains in first person and second person are carried entirely by verb endings. This is the case for fourteen of the twenty-two first person chains and seven of the fourteen second person chains. By contrast, only two (in 7:39 and 15:7) of the twenty-one third person chains are carried entirely by verb endings, and in both of these cases, the first reference in the chain resumes a subject which was previously mentioned by a noun. In 7:39 the noun occurs in the previous colon, and in 15:7 the noun occurs at the beginning of the chain begun in 15:3, which subject is resumed in 15:7.

Perhaps the most unusual third person chain begins in 15:42 extending across four colons and filling seven (or perhaps eight, if the two bodies are seen as identical) subject slots. Here the ψυχικόν 'soulical' body which dies is contrasted with the spiritual body which will be raised. But the noun σω̑μα 'body' is postponed until the end of the chain, and the chain is carried at first by verb endings.

All of the other third person chains begin with either a noun (in eleven cases), a pronoun (in five cases), a clause (once, in 15:36), or an article (once, in 10:12). This is in sharp contrast to first person chains (where only six chains begin with pronouns) and second person chains (where only four chains begin with pronouns and only one with a noun). The last mentioned begins in 1:26 with a vocative noun ἀδελφοί 'brothers', referring to the Corinthians, and continues with three uses of the third person πολλοί 'many'.

The other grammatical device worthy of note for carrying a same subject chain is a series of pronouns. The second person case begins in 1:12 with the phrase ἕκαστος ὑμω̑ν 'each of you' and continues for four subject slots with first person singular pronoun ἐγώ 'I'. The third person examples are found in 10:23 and in 12:29, repeatedly using the words πάντα 'all things' and πάντες 'all people', respectively.

Finally, the relation between subject tracking tasks and word order within a clause needs to be explored. Table 28 gives the word order within clauses for noun subjects as distributed among tracking tasks. The table has been divided into three general areas: word orders in which the subject precedes the verb, word orders in which the subject follows the verb, and word orders without a verb. The totals for all three of these areas do not show a statistical significance for the distribution, but the difference between the first two areas is statistically significant.

TABLE 28

RELATIONSHIP OF CLAUSE ORDER AND TRACKING TASKS

FOR SUBJECT NOUNS IN I CORINTHIANS

                                                                   
| Clause | Continuation | Resumption from | Different  | New       |
| Order  | of Same Sub. | Clause or Colon | Reference  | Reference |
| SV     |            6 |            9    |         51 |         1 | 
| SVO    |            3 |            4    |         29 |         1 |
| SOV    |            3 |            5    |         48 |         4 |
| OSV    |            0 |            0    |          4 |         0 |
| Totals |   (34.3%) 12 |   (43.9%) 18    |(50.4%) 132 | (31.6%) 6 | 
| Verbal |   (48.0%)    |   (56.3%)       |(71.7%)     | (46.2%)   | 
| VS     |            7 |           13    |         22 |         3 | 
| VSO    |            3 |            0    |          4 |         3 | 
| VOS    |            0 |            0    |         10 |         1 | 
| OVS    |            3 |            1    |         16 |         0 | 
| Totals |   (37.1%) 13 |   (34.1%) 14    |(19.8%)  52 | (36.8%) 7 | 
| Verbal |   (52.0%)    |   (43.7%)       |(28.3%)     | (53.8%)   | 
| SO     |            5 |            1    |         11 |         3 | 
| OS     |            1 |            1    |         21 |         0 | 
| other  |            4 |            7    |         46 |         3 | 
| Totals |   (28.6%) 10 |   (22.0%)  9    |(29.8%)  78 | (31.6%) 6 |

 

The data for new and different reference have been split into separate columns for Table 28 because these show quite different results. New references are those topics that are introduced for the first time in the letter at this point. Different references are those that are not new to the letter but are different from those topics used as a subject in the immediately preceding context; they include both topics that have been introduced in an object slot and those that are resumed from a subject slot found at least two colons previously.

Omitting for the moment the SO and OS clauses and just looking at clauses with verbs, 71.7% of the nouns that introduce a different subject than one recently discussed are found in an SV word order, while only 46.2% of the subject nouns that introduce a new topic to the letter are found in this order. There is perhaps a correlation between the fact that a majority of new topics are introduced in an object slot and over half of those introduced in a subject slot follow the verb, but that would require further work in object slot participant tracking which is beyond the scope of this study. It is noteworthy that among those clauses that introduce new topics in the subject slot, only one has the object preceding the subject. Again, in clauses with verbs the majority of the subjects which continue the same topic (52.0%) are also found in a VS word order, while 48.0% have an SV word order. Conversely, 56.3% of those subjects which resume a recently discussed topic favor an SV word order, while only 43.7% have a VS word order. It seems likely that resumption favors subject fronting, with the probability of fronting increasing as the distance to the previous discussion of the topic increases.

Back to top | Next: Chapter 5.3 Clause Word Order


Bruce Terry's Home Page
Bruce Terry Home Page
http://bible.ovc.edu/terry/dissertation/u-5_2-part.htm
Last Updated December 1, 2003
Page developed by Ben Cheek and Paul Rennix and maintained by .

Copyright © 1993 Ralph Bruce Terry. All rights reserved.