Using Linguistic Features to Analyze Romans 5: An Application of the OpenText.org Discourse Model
by Sean Adams and Philip Burggraff (02/23/2006)
This brief paper is designed to provide one application of the OpenText.org model to a biblical passage in order that those who are new to this site or new to the concept of evaluating a text using clausal relationships might better understand how to use this site and be exposed to some of the possible benefits that this approach might have. This study uses the functional display and the clausal display models from the OpenText.org site and attempts to analyze Romans 5 according to the features of clausal relationships, verbal aspect, participant reference, theme and semantic domains. The paper discusses each of these features individually, beginning with an introduction that explains what the feature is and how it works within the OpenText.org model and followed by application of the feature to Romans 5. It concludes with a discussion of the overall flow of the passage.
According to the OpenText.org discourse annotation model, the way in which clauses relate to each other belongs to the discussion of the paragraph level, and OpenText.org is in the process of developing a system to analyze this level. Yet, in order to show some level of relationship between clauses, OpenText.org has "included some basic connection and dependency information at the clause level."1
OpenText.org uses three types of clauses in their annotation scheme and labels all clauses according to them. These types and the formal/categorical features that determine them can be seen in the following chart.
OpenText.org has also labeled the connections between clauses. At the beginning of any clause, that particular clause is given a label regarding which number clause it is in the chapter of text it is found. Underneath this number is found a reference to another clause to which that particular clause most immediately relates. While OpenText.org has not yet attempted to categorize these relations as to the function in the text (such as purpose, reason, cause, condition, etc.), seeing the flow of the text according to primary, secondary, and embedded relationships is important in understanding how the writer introduces new information and moves the message of the text forward (through primary clauses), as well as further defines and adds to information already present in primary clauses (through secondary and embedded clauses). 2
Application to Romans 5 3
Romans 5 begins with a primary clause (c5_1) stating "we have peace with God through our Lord Jesus Christ". The circumstance or reason that we have such peace, "being justified by faith," is given in the embedded clause (c5_2) in the adjunct at the beginning of the primary clause. A secondary clause further describes what "we have obtained" through Jesus Christ. The primary clause c5_5 continues the discussion from c5_1 by adding, "we exult in hope of the glory of God." Both c5_6 and c5_7 are primary clauses and further move the discussion forward, stating that "we" glory not only in hope but also in "our tribulations". Clauses c5_8-c5_14 are a series of secondary and embedded clauses that describe what "we" know, which leads to glorying in tribulations. In these opening clauses, the subject in the primary clauses has consistently been the "we" from the first person verbs that have dominated these clauses. This consistency in subject shifts in relation to the primary clauses, beginning with clause c5_15.
The primary clause c5_15 states what Christ has done for "us": while "we were helpless," he "died for the ungodly." Clauses c5_17 - c5_18 provide an illustration showing the extreme actions that Christ went to in dying for the ungodly by pointing out that someone might possibly die for a righteous or good man, but even that is doubtful. Primary clause c5_20 relates how God has demonstrated his love for "us" and states this demonstration in the secondary clauses c5_21 - c5_22, "while we were sinners, Christ died for us." The primary clause c5_23 states, "as we are justified by his blood" (embedded clause c5_24), we will be saved from God's wrath through the same one who justified us. This is further substantiated by primary clause c5_27 and its secondary clause c5_25, in which the author states that if "we", being enemies, are reconciled to God by his Son's death, "we" will certainly be saved by his life. Primary clauses c5_29 and c5_30 add again that "we" glory (cf. c5_5 and c5_7), yet this time the glorying is in God through our Lord Jesus Christ, through whom "we" have received reconciliation (secondary clause c5_32).
Clauses c5_33 - c5_44 are a series of secondary clauses that lead to the primary clause c5_45. These clauses reveal the impact of sin and death and mention Adam as a type of the one to come. Primary clause c5_45 states that the gift is not like the transgression. Secondary clause c5_46 and primary clause c5_47 contrast the onset of death to the many and the grace and gift abounding to the many. The contrast continues into primary clauses c5_48-c5_51 with judgment leading to condemnation but the gift out of transgressions leading to justification. Clausal pairs c5_52 and c5_53, c5_55 and c5_56, and c5_57 and c5_58 consist of a secondary clause followed by a primary clause to further contrast the one man's transgression and disobedience and their results (death, condemnation, and many sinners) with the other man's abundant gift, righteousness, and obedience and their results (reigning, justification, and righteousness). Primary clause c5_59 and primary clause c5_61 and their respective secondary clauses bring the contrast to a conclusion by stating that the Law's coming led to transgression increasing, but where sin increased, grace increased more leading to eternal life through Jesus Christ.
In conclusion to this section, the flow of the passage is seen by tracing the primary clauses in Romans 5. The chapter begins with a recitation of what "we", having been justified, have, namely peace with God, and what we can glory in, specifically hope of the glory of God and also tribulations. The text then discusses what Christ and God have done for "us", which leads again to a discussion of what "we" will have, salvation from wrath by His life, and can glory in, namely God. The text then moves to a lengthy contrast between Adam's sin, death, and condemnation and Jesus Christ's gift, righteousness, and justification to life. While the Law led to the increase of sin, grace increased even more, ultimately resulting in eternal life for those who had received the gracious gift of Jesus Christ.
Verbal aspect, contrary to tense-based models, proposes that verbs in ancient Greek do not incorporate a literal time reference. Rather, verbal aspect is a semantic category by which a writer represents a perspective on an action by grammaticalizing it through a selection of a particular tense-form. Accordingly, verb tense-forms do not inform the reader regarding temporal relationships, which are incorporated through larger grammatical and conceptual units, but, through the choice4 of a specific tense-form by the author, inform the reader of the relative importance that the action or the section has as a whole.5
Within the verbal system, aspect is divided into three categories: perfective (aorist), imperfective (present and imperfect) and stative (perfect and pluperfect).6 In their functional display model, OpenText.org identifies these aspects with corresponding colours: blue (perfective), yellow (imperfective) and red (stative). These categories represent the different levels of emphasis that a writer could impress upon the reader. The least marked is the perfective aspect because it is the most commonly used within the New Testament. Accordingly, the perfective aspect provides the default tense or background information within both narrative and expositional passages and the backbone within narrative passages. The imperfective aspect is slightly more marked and is used to create emphasis within a passage. This aspect is also used to form the backbone within expositional passages; however, it still maintains its markedness in comparison to the perfective aspect, which provides the background information. The stative aspect is the most marked and is utilized to highlight important themes and events by the author. Overall, the author chooses the particular verbal aspect according to his or her understanding and interpretation of the events. It seems best, therefore, to consider these aspects as contributing to the prominence of a particular theme or passage, which is being projected onto the text at varying discourse levels. Consequently, this is where the grammatical interpretation of the aspect should begin. 7
Application to Romans 5
In evaluating the aspectual usage within Romans 5, two sections emerge: vv. 1-11 and 12-21. In the first section there are 14 occurrences of verbs in the primary clause. 8 Out of these 14 verbal instances, 4 are perfective, 7 are imperfective and 3 are stative. 9 This distribution is significant because, typically, perfective verbs are dominant and occur most frequently. The fact that a majority of the verbal aspect choices are imperfective, with three stative aspects, signifies that this passage is important to the author and causes it to be a marked section. In addition, within this section, there are possibly different emphases. For example, in vv. 1-5 five of the eight verbs are imperfective, and the other three are stative. Consequently, vv. 6-11 are composed using a majority of perfective aspect verbs. This forms a high concentration of markedness within the verses 1-5, which might indicate an authorial emphasis.
Conversely, Romans 5:12-21 also contains 14 verbs, however, 13 of them are perfective, with one imperfective and no stative aspect verbs. As a result, this passage would be considered not marked because of the consistent use of the default perfective aspect. This creates a marked discontinuity between these two sections and indicates to the reader that the author might be moving to a new section or theme.
In addition to verbal aspect, discourse analysis theory relies heavily on other features of the text to develop prominence, such as participant reference and semantic domains. Reed rightly states that prominence in discourse 'is rarely signaled by one device, but more often is the result of a combination of grammatical and semantic features'.10 Consequently, this study turns to an evaluation of participant reference.
While the feature being analyzed in this section is that of participant, the notion of participant can only be understood in congruence with the notion of process. As Halliday notes, processes consist of the "goings-on" (doing, happening, feeling, and being) of one's conception of reality. 11 Participants are the ones involved in the process being described. In language, there are a number of processes that exist, and participants take on different roles within the various processes. These processes and the relation of participants to them has been summed up well in the following chart by Reed. 12
Figure 2. Jeffrey Reed's table describing process types and their participants.
OpenText.org traces the various participants involved by marking each time a first, second, or third person reference is made within a clause. In certain instances the participant is overtly stated in the subject or component word group, but in other instances it may simply be found at the end of the verb form. Tracing participant reference can be a crucial step in noticing shifts within a discourse.
Application to Romans 5
While the sections on clause connections and verbal aspect have already introduced participant reference, a more thorough accounting of the participants is necessary in order to draw any conclusions. Clauses c5_1 - c5_7 utilize first person reference exclusively, which is seen in both 1st person verb forms as well as pronouns. The subject of the mainline verbs is never explicit but is found in the 1st person verb forms. The secondary clauses that follow (c5_9 - c5_13) utilize third person reference through a number of third person subjects - θλῖψις, ὑπομονὴ, δοκιμὴ, ἐλπὶς, and ἀγάπη. Yet, these are secondary clauses to what preceded in which first person reference dominated; therefore, the first person references carry these opening clauses forward.
In clauses c5_15 - c5_22 this first person dominance is replaced with third person references. The subject of the primary clauses is explicitly stated in all four primary clauses of this section -- Χριστὸς, τις (2x), and θεὸς. While the third person references have taken over as subject of the mainline verbs, first person reference is still found in the embedded clause c5_16 (ἡμῶν as genitive subject) located in primary clause c5_15 and the adjunct of primary clause c5_20 (εἰς ἡμᾶς). Thus, third person subjects interacting with first person referents carry this section forward.
c5_23 - c5_32 revert back to first person reference as the subject of the primary clauses, but the verbs of these clauses are mainly passive. Again, the first person references are seen in the first person verbal forms, as well as the pronoun ἡμῶν in the embedded clause c5_31. None of these clauses contain an overtly stated subject. While the subject is first person, this subject is the receiver (goal) of the actions stated in the verbs. The actor doing the verbs is Jesus Christ as seen from the explicit reference to him in clause c5_30 and the 3rd person pronouns of clauses c5_23, c5_25, and c5_27.
Throughout clauses c5_33 - c5_64, the participant reference is almost entirely third person, with all finite verb forms in these clauses found in the third person. This holds for both the primary and the secondary clauses in this section. In almost every clause, the subject is explicitly stated. The only exceptions are the primary clause c5_56 and the secondary clauses c5_44 and c5_55. The only first person reference is found in the secondary clause c5_64 at the end of the section with the pronoun ἡμῶν in the final adjunct of that clause. Thus, this section is dominated by third person participant reference.
In regards to participant reference, Romans 5 displays a stark contrast between its beginning and its ending. There is both first and third person participant reference in clauses c5_1 - c5_32. First person references carry the primary clauses along from c5_1 - c5_13. They fill the roles of carrier (possessor of peace) and actor. Third person referents interact with first person referents in the primary clauses of c5_15 - c5_22. The role of actor, though, is filled by the third person referents (mainly Jesus Christ and God). In clauses c5_23 -c5_32, first person referents fill the subject slot again, but the role of actor continues to be the third person referents (Jesus Christ). In contrast to this section, third person participant referents exclusively carry forward the primary and secondary clauses of c5_33 - c5_64. The only first person reference is the use of a pronoun in the final clause of the section.
OpenText.org tracks information at the clause level by noting what element in the clause is carrying forward the theme of the clause. At the clause level, the theme is the starting point for the message, and it reveals what the clause is going to be about. 13 In languages with flexible word order, the theme naturally comes in the first position of the clause. 14 The remainder of the clause in which the theme is developed is labeled the rheme. 15 At the clause level, theme does have a very useful function in that "it signals the point at which the information carried by the clause attaches to the preceding discourse, it provides cohesion." 16 In OpenText.org's annotation, the functional chart marks which component (subject, predicator, complement, or adjunct) comes first in the clause and thus signals the theme of the clause.
Application to Romans 5
Throughout Romans 5, the theme of the majority of clauses is introduced by the adjunct. Of the twenty-two primary clauses found in Romans 5, the adjunct introduces theme in fifteen of them. The only exceptions, though, do reveal an interesting pattern. The first three primary clauses that have their theme introduced by a component other than an adjunct are c5_5, c5_7, and c5_20. The predicate component introduces the theme in all three of these clauses. The next primary clause with a non-adjunct introduced theme is clause c5_30. In this case the theme is introduced by a complement component. The final three primary clauses with their theme not introduced by an adjunct are c5_50, c5_51, and c5_59, in which case a subject component introduces all three of these clauses. These occurrences do lend themselves to the segmenting of this text into two distinct sections. The three clauses with their themes introduced by a predicate component are all found in the first section (c5_1 - c5_32). The one clause with its theme introduced by a complement is the final primary clause of the first section, thus standing at the boundary of the two sections. The final three clauses with their themes introduced by a subject component are all found in the final section (c5_33 - c5_64).
Semantic domains are words that are grouped together because they all have shared semantic features. In Domain 19 Physical Impact, for example, κολαφίζωa (19.7), ῥαβδίζω (19.8), and μαστίζω and μαστιγόωa (19.9) all share the features of physical impact involving hitting or striking. They differ, however, in certain distinctive features in that κολαφίζωa designates striking or beating with the fist, ῥαβδίζω designates beating or striking with a stick or rod, and μαστίζω and μαστιγόωa designate beating with a whip. 17 Within the OpenText.org clause and domain display, the domains that are associated with each word are given. In order to streamline some of the information, primary domains are given and tallied within the functional display model. This provides the user with a visual and graphical representation of semantic occurrences.
Semantic domain theory, mainly developed by Louw and Nida, posits the idea that words with similar semantic meanings, which are grouped in common domains, are related to each other and can be used to form coherence within a text. 18 This allows the scholar the ability to explore general semantic patterns that occur across a section of text without being fixed to a single word or cognate. 19 This principle is important for the study of discourse analysis because it provides an additional method of examining linguistical patterning that goes beyond traditional word counts.
An important aspect of developing a discourse is creating cohesion within a passage. One way to accomplish this is by the repetition of words that help create connections within a passage and help the discourse 'hang together' internally. 20 In a discourse, a string of semantically related words is combined to form cohesive ties, which are used by an author to create links to preceding texts that help to create unity within a passage. Halliday defines cohesive ties as '…relations that may involve elements of any extent, both smaller and larger than clauses, from single words to lengthy passages of text; and that may hold across gaps of any extent, both within the clause and beyond it, without regard to the nature of whatever intervenes'. 21 Overall, lexical cohesion appears as the most frequent cohesive tie within a discourse. 22 These tools are important for determining the boundaries of a passage and the relative unity of that passage.23
Application to Romans 5
In evaluating semantic domains it is important to identify any groupings of domains, because these collections might indicate a common theme or an emphasis of the passage. In determining the markedness of a passage, it is important to determine the linguistic prominence of particular semantic domains. A high concentration of a domain, with few other uses increases the prominence of a section and contributes to the development of topic. The first grouping that is apparent within Romans 5 occurs between verses 2 and 8 where there are nine occurrences of semantic domain 25 'Attitudes and Emotions' (ἐλπίς, ἀγάπη, καυχάομαι, ὑπομονή). Also significant is the fact that there are no other occurrences of domain 25 anywhere else in Romans 5. This lends credence to the idea that domain 25 might create coherence within the first section.
The second semantic grouping occurs between verses 15 and 19 where there are 22 occurrences between the related semantic domains of 59 'Quantity' and 60 'Number.' In this section the author switches back and forth between εἷς (one) and πολύς (many) creating movement within the text between Adam and Jesus, the two individuals who affected the world and the many, those who were affected by their actions. It is through this interplay between the one and the many that Paul develops his rhetoric and creates solidarity within the passage and distinguishes it from the previous section.
The final marked semantic occurrence begins at verse 12 and continues to the conclusion of the chapter. This forms a semantic boundary, which is a point in the discourse where a shift in semantic domains occurs. This shift, which is usually accompanied by other additional discourse changes, creates a 'zone of turbulence' and might signify a discourse peak.24 In this section there are 22 occurrences of domain 88 'Moral and Ethical Qualities and Related Behavior' with at least one occurrence in every verse. The repetitive use of ἁμαρτία, παραπτώμα, χάρις, and δίκαιος all focus the reader's attention on the ethical and behavioral issues. The consistent repetition of this domain forms a strong semantic tie within the section and creates coherence of topic and focus.
Through the use of semantic domains and ties the division between vv. 1-11 and 12-21 are solidified. When taken in relationship to participant reference the topic of the section is revealed. According to Westfall, 'a topic above the sentence level is determined by one of the following or a combination of the following criteria: tracing the participant and process semantic chains and their interaction…'25 By identifying the sharp change in participant reference from alternating first and third person, to only third person references and by pairing this information with the development of semantic domains, it appears that the topic of this second section is the moral behavior of people, in light of the actions of Adam and Jesus.
The analysis of the five features revealed a strong argument for breaking this text into two main sections (Romans 5:1-11 and 12-21) with the first section segmenting into two sections (vv. 1-5 and vv. 6-11).
Tracing the clause connections reveals that Romans 5 begins by stating what "we" have and can boast in (c5_1 - c5_13). It then moves to what Christ and God have done for "us," and what we will have because of their action (c5_14 - c5_32). The second half (c5_33 - c5_64) discusses the contrast between Adam, sin, death, and condemnation and Jesus Christ, the gift, grace, righteousness, and justification. The section concludes that while the Law led to the increase of sin, grace increased even more.
The evaluation of verbal forms in Romans 5 further emphasizes the break between vv. 1-11 and vv. 12-21. This is suggested by the high concentration of imperfective and stative aspect forms found within the first half, followed by the return to the perfective aspect and its continuation throughout the remainder of the passage.
The analysis of participant reference reveals a progression. The chapter begins with first person referents at the fore (c5_1 - c5_13), moves to interaction between first person and third person referents (c5_15 - c5_32), and concludes with third person referents dominating the primary clauses (c5_33 - c5_64).
In discussion of theme, while adjuncts are dominant throughout the passage, predicates and a complement did introduce theme in the first half of the chapter (c5_1 - c5_32). In contrast, the only complement to introduce theme in the second half (c5_33 - c5_64) other than adjunct was the subject component.
In addition to this, the distribution of semantic domains provides strong support for this break. In the first section there is a grouping (nine occurrences) of domain 25, which is notably absent in the second section. Likewise, in the second section, there is a strong cohesive tie with the repetition of domain 88 'Moral and Ethical Qualities and Related Behavior.' This tie creates coherence within the passage and distinguishes itself from the previous section. In addition to this, there are a large number of instances (22 within verses 15 and 19) of the semantic domains 59 'Quantity' and 60 'Number,' which creates a strong sense of coherence within the passage.
The occurrence of one of these features might not be significant, however, the combination of them forces the reader to acknowledge that there is a separation between verse 11 and 12 and that verses 1-11 are marked in their use of verbal aspect and participant reference.
1M.B. O'Donnell, "Introducing the OpenText.org Syntactically Analyzed Greek New Testament", 22 September, 2005, http://divinity2.mcmaster.ca/OpenText/resources/articles/a8
2For further discussion of the flow of the text through the use of primary and secondary clauses, see O'Donnell, 'Introducing OpenText.org'.
3The discussion that follows will utilize and make reference to both the Clause Annotation of Romans 5 display and the Functional Clause display of Romans 5 from the OpenText.org website. The Clause Annotation of Romans 5 display presents the easier way to view primary, secondary, and embedded clausal relationships. The Functional Clause display of Romans 5 traces a number of occurrences of various linguistic phenomena throughout the passage.
4One of the basic tenets behind the theory of verbal aspect is that the author makes a conscious choice regarding the verb form that they use. Some scholars question this, claiming that an author is incapable of holding all of these considerations in their head at one time. This might be the case, however, one must take into account the idea that a native language user will have internalized a number of these methods making them naturally flow into their writing. Nevertheless, it is still possible for them to choose, as strategic times, to incorporate different verb-tenses.
5S.E. Porter, Idioms of the Greek New Testament (BLG 2; Sheffield: Sheffield Academic Press, 1992), pp. 20-29.
6It is important to note that the future tense-form does not represent a time-based or a verbal aspect tense-form, and consequently, is not incorporated into this discussion. For further information see chapter nine in S.E. Porter, Verbal Aspect in the Greek of the New Testament, with Reference to Tense and Mood (SBG 1; New York: Peter Lang, 1989).
7 Porter, Verbal Aspect. See chapter two.
8In this study, and within the functional display model within OpenText.org, only verbs that occur within primary and secondary clauses are evaluated. Although this does not take into account occurrences within subordinate or embedded clauses, it places more weight on verbs that govern the sentence and subordinate clauses. Typically primary clauses convey the most prominence in discourse, while secondary clauses increase dependence and, in turn, decrease the level of importance. See S.E. Porter, "Prominence: An Overview" in S.E. Porter and M.B. O'Donnell (eds.), The Linguist as Pedagogue: Trends in Teaching and the Linguistic Analysis of the Greek New Testament (NTM 11; Sheffield: Sheffield Phoenix Press, forthcoming).
9There is some debate concerning the optimal method of determining significance within a discourse feature. The standard measuring technique is to note occurrences per hundred or thousand words. However, an alternative method, described by Ball, posits that the occurrence of a feature should be measured against the number of possibilities for it to occur. This has particular merit because, in this case, sections of lists or large subordinate or embedded clause sections do not skew the results of verb occurrences. C.N. Ball, 'Automated Text Analysis: Cautionary Tales', Literary and Linguistic Computing 9.4, pp. 295-302, (297).
10J.T. Reed, A Discourse Analysis of Philippians: Method and Rhetoric in the Debate over Literary Integrity (JSNTSup 136; Sheffield: Sheffield Academic Press, 1997), p. 111.
11M.A.K. Halliday, An Introduction to Functional Grammar (2nd edn; London: Arnold, 1994), p. 101.
12Reed, Discourse Analysis of Philippians, p. 69.
13Halliday, Functional Grammar, p. 39.
14Halliday, Functional Grammar, p. 39; Reed, Discourse Analysis of Philippians, p. 103; K. Callow, Discourse Considerations in Translating the Word of God (Grand Rapids: Zondervan, 1974), pp. 58-9.
15Halliday, Functional Grammar, p. 38 and Reed, Discourse Analysis of Philippians, p. 103.
16Callow, Discourse Considerations, p. 58.
17J.P. Louw and E.A. Nida, Greek-English Lexicon of the New Testament Based on Semantic Domains (2 vols.; 2nd edn; New York: United Bible Societies, 1989).
18Louw and Nida, Greek-English Lexicon.
19Reed, Discourse Analysis of Philippians, pp. 76-78.
20M.A.K. Halliday, and R. Hasan, Cohesion in English (English Language Series; London: Longman, 1976), pp. 4-5; M.A.K. Halliday, and R. Hasan, Language, Context and Text: Aspects of Language in a Social-Semiotic Perspective (Geelong, Australia: Deakon University, 1985), p. 48
21Halliday, Functional Grammar, p. 309.
22M. Hoey, Patterns of Lexis in Text (Oxford: Oxford University Press, 1991), pp. 3-10.
23For futher examples on the use of semantic domains see: M.B. O'Donnell, Corpus Linguistics and the Greek of the New Testament. (NTM 6; Sheffield: Sheffield Phoenix Press, 2005); Reed, Discourse Analysis of Philippians; and C. Westfall, 'Blessed Be the Ties that Bind: Semantic Domains in Hebrews 1:1-4:16', presented at SBL 2005.
24R.E. Longacre, "Discourse Peak as a Zone of Turbulence," in J.R. Wirth (ed.), Beyond the Sentence: Discourse and Sentential Form. (Ann Arbor: Karoma, 1985), p. 85.
25Westfall, 'Blessed Be the Ties that Bind', p. 12.
Discuss this article in the OpenText.org discussion forum