Ki/gar and waw/kai are often markers and not words to be translatedRelated Media
Pastor of Woodlyn Community Chapel, Ferndale, WA
formerly a Translation Consultant with Bibles International
delivered at the NW Regional ETS Meeting, February 27, 2010
1. An introduction to the problem
Many verses in modern English Bible versions still begin with “And,” “But” or “For.” This practice needs to be reevaluated by taking a look at these words as discourse markers especially when they are found in initial sentence position. This is important for a number of reasons. For instance, literacy concerns focus not only on form but also on function. Readability raises various interesting questions, such as sentence length, the essence and form of a sentence, and the clear identification of subject and verb. Resolutions for all of these issues help us to form better conclusions about the way Scriptures should be translated.
The most ancient copies of the Hebrew and Greek texts of the Bible did not contain common punctuation markers, which for us help in the ease of reading. It is the thesis of this article that these texts contained words which served as indicators or markers of the larger discourse unit. These served not only as quasi-punctuation marks but also as dependency indicators between linguistic clauses or sentences, binding them into a whole. Most of these markers continue to function in the same way in oral speech today. In written form, they are sometimes better left untranslated in English, or as the case may be, analyzed for translation to see what kinds of markers they are, and then what words should be used to render them. Most often, a sensitive translation will not leave us with the standard “For” or “And.”
Incidentally, this issue raises questions about the viability of sentence-only grammars. What constitutes a sentence? Where does it begin and end? Can all sentences, typically understood as such, stand alone, or do a good proportion of them only make sense in a context? Do proper connectives, such as “therefore,” “so” and “thus,” keep us on track for the formation of a complete sentence? Likewise, does the use of “also,” “in addition,” and “moreover” serve any better than a simple “and” in keeping us from being dependent on a preceding statement? The answer is no. So why be concerned about the overuse of “And” or even the usage of “For” to start sentences? Is it only about variation in style?
The typical word-for-word translation focuses on words alone. One could argue that the New King James Version of the Bible [NKJV] is closer to the text of Romans than the New International Version [NIV] merely by counting the number of times one has translated gar in the Greek to “for” in the English. The argument goes this way—each gar is a significant indicator. To eliminate one of them tears the context apart.
However, one may also argue that repetitiveness of any one word not normally occurring in a target language is annoying. Sometimes, one more connector in a paragraph and/or linguistic context gets in the way of the argument. This was evidently not true in the source language, but it may be in the target language. So, the important question is, “Do we properly understand the original use of these connectors or indicators of dependency as they were used in the source language?”
2. Some basic literacy concerns
Translators of first-time tribal language Bibles are concerned about the needs of their readers and hearers. How well educated are they? What is their reading level? Generally speaking, one can often expect a readership level at about the third grade for many of these readers. Sentence length should be at an average of thirty words, unless providing for a long list of items. Sentences should be complete and well connected within the context. Punctuation marks should restrict the use of semi-colons with the single exception of separating larger items within a long list of items. This is so because the semi-colon can act as either comma or period and tends to confuse the reader.
3. What is a sentence?
Implicit within our discussion is the definition of a sentence. A sentence is the smallest unit of grammar that voices a complete thought and has the potential to stand alone. Allerton summarizes the history of the notion of a sentence:
“Traditional attempts to define the sentence were generally either psychological or logical-analytic in nature: the former type spoke of 'a complete thought' or some other inaccessible psychological phenomenon; the latter type, following Aristotle, expected to find every sentence made up of a logical subject and logical predicate, units that themselves rely on the sentence for their definition. A more fruitful approach is that of Jespersen (1924: 307), who suggests testing the completeness and independence of a sentence, by assessing its potential for standing alone, as a complete utterance.”1
The above quote from Jespersen is here: “A sentence is a (relatively) complete and independent human utterance—the completeness and independence by its standing alone or its capability of standing alone, i.e. of being uttered by itself.”2 From this, we understand that dependent clauses do not make sentences. Also, dependent clauses must refer back to the main clause of the sentence in which both are found.
We recognize that a sentence will not express a thought absent from the context in which it is uttered. This is why we notice so many transition words and connectors, whether conjunctive or disjunctive, within a paragraph, or a larger unit of discourse. These are for the purpose of providing cohesion, indicating a relationship of dependency from sentence to sentence. The use of the semi-colon has a comparative relationship with the use of initial sentence-position markers. A writer often uses the semi-colon instead of a period when seeking to emphasize a close or special relationship between two sentences.
4. Proper and improper uses of initial position conjunctions
Let us look at the frequent use of “And” as well as “But” to begin sentences and paragraphs. This can occur at many levels—in popular journalistic writing and also in certain academic journals. Both The Shorter Oxford English Dictionary and Webster’s Unabridged Dictionary allow for the use of “and” to begin sentences.3 For the use of “but,” the Shorter Oxford English Dictionary has: “In a simple sentence, introducing a word, phr. or (rarely) a clause…”4 Webster’s has: “…sometimes used at the beginning of a separate sentence.”5 Neither dictionary mentions or gives an example of the conjunction “for” beginning a sentence. The usage is accepted for the use of “and,” while the use of “but” is somewhat restricted. Some readers will identify with the feeling that these connectors are overused. However, does this reflect the oral use of language of only certain speakers? More importantly, does its usage annoy us and why?
Here is an example of what one may consider an overuse of “And” and “But.”6 In an academic article of a journal respected by many, the author, well-educated and widely published, used these two connectors 79 times to start sentences within an article of 17 pages long. He started three paragraphs with “And” and six paragraphs with “But.” By the time one noticed all these connectors, he or she may have lost the flow of the article’s argument. One could argue, however, that the Bible does the same thing on an even greater scale. It is for this reason that we should be concerned about these transition markers and how best to represent them in the text.
5. Waw/kai in sentence initial position
The particles waw/kai in the Bible serve as both connectors and markers. Not only do they indicate the beginning of independent units of thought, they also indicate coordination and continuance of the context, whether the topic is explanatory, causal, or temporal. This is why these particle connectors could occur at the beginning of dependent clauses, and also at the beginning of independent clauses and sentences, as well as at the beginning of large units of discourse, such as paragraphs, chapters (Isaiah 3:1), and even books (Exodus 1:1).
“Waw often appears at the beginning of a sentence as an introductory particle with a meaning such as ‘then.’ The book of Exodus actually begins with such a waw, but whereas KJV and NASB translate it by “now,” other versions, such as RSV and NIV, leave it untranslated.”7 The exegetical waw is another usage that can introduce apposition to explain or add emphasis (1 Samuel 2:2, Isaiah 40:10, 44:1). English equivalents might be “such as, that is, for example, and now, then, which leads us to, etc.”8
In many cases, waw/kai does not need to be translated. In these cases, it is a marker of continuation, a binder that provides cohesion. If it is not understood as such in a target language, a literal representation in translation confuses rather than clarifies. Louw and Nida tell us that all markers of transition, including γαρ, και, ἀλλα, ὑμεν, νυν, δε, γινομαι, are often best left untranslated. In addition, και and γαρ are often to be understood as markers of a new sentence.9
A. T. Robertson concurs from a historical perspective on the use of connector particles as cohesive facilitators.
“The Greeks, especially in the literary style, felt the propriety of indicating the inner relation of the various independent sentences that composed a paragraph. This was not merely an artistic device, but a logical expression of coherence of thought. Particles like και, δε, ἀλλα, γαρ , ουν, δη, etc., were very common in this connection.”10
Steven Runge examines New Testament Greek grammar with the linguistic aid of discourse theory. This approach attempts to go beyond the traditional sentence-only grammar, and as a result reaps benefits when examining the phenomenon of particles. He warns us that English is more apt to use particles sparingly, that is when compared with NT Greek [or for that matter OT Hebrew].11 Asyndeton refers to the lack of a conjunction to link clauses. Asyndeton is the English default connective, signifying that the clause relationship is sufficiently understood.12 On the other hand, the Greek particle καί associates two thoughts more closely than no conjunction at all, usually connecting two equal thoughts in the mind of the writer.13
Occasionally, versions of the Bible will translate waw with the word “for.” The following Hebrew grammarians do not mention waw used as an explanatory conjunction: Joüon, Williams, and Waltke and O’Connor. However, Gesenius listed several examples found in Psalm 60:13, Genesis 6:17, 22:12, Exodus 23:9, Job 22:12 and perhaps Psalm 7:10.14 Of these examples, only one begins a sentence and that is Genesis 6:17 (RSV, notice that the NIV excludes the “For” here). “For behold, I will bring a flood of waters upon the earth, to destroy all flesh in which is the breath of life from under heaven; everything that is on the earth shall die.”
6. ki/gar in sentence initial position
A recent New Testament grammar provides us with little help here, as it chose to focus on the sentence as its highest form of syntax. The author, Daniel Wallace, listed the particle γάρ as both a coordinating and subordinating conjunction.15 Later on, he also listed it as an explanatory (for, you see) and an inferential (therefore) conjunction.16
Let us first look at its Old Testament Hebrew counterpart ki, for which more data can be found. Harman defines its usage as one that introduces clauses to show connection between two of them or to show emphasis.17 Another initial position use includes oaths that can be preceded or introduced by “ki,”18 as well as the emphatic or deictic use which many believe to be the particle’s underlying meaning. In Albrektson’s anthology of Old Testament studies, Schoors states that ki is first and foremost a deictic and demonstrative particle, as it directs the path of the discourse.19 Joüon tells us that it begins sentences of affirmation, and may or may not need to be translated. The sense is “Yes,” or “Certainly.” It can be found in oaths and also in the apodosis of the conditional sentence. Examples are found in Genesis 18:20, 1 Kings 1:30 and perhaps also in Isaiah 32:13, Psalm 49:16; 77:12; 118:10; and Lamentations 3:22.20
Waltke and O’Connor write:
“Traditionally ki is considered a conjunction (cf. ‘for’), but we consider it to be an emphatic adverb (cf. ‘indeed”). The question is not primarily one of translation (though the standard translation ‘for’ is sometimes illogical and often tedious), but rather of aligning ki with other forms that work similarly.”21
Commenting further on the translation difficulties of ki, Waltke and O’Connor add that it is the most problematic of the following four adverbs beginning with k—ken, kah, kekah, ki. The emphatic and the logical are the two clause-adverbial uses of ki. The logical use
“overhadows the first [the emphatic] through the dominance of the translation ‘for’ in ‘Biblical English.’ This translation is often used where it, and the understanding behind it, are simply wrong, that is, where there is no evident logical link of the clause to what precedes. Further, ‘for’ suggests that ki is a subordinating conjunction, which it often is not when used in the logical sense.”22
Follingstad goes one step further when he declares the particle ki to be “discourse deictic.” This refers to the “function of ‘mentioning’ propositional content – a thought implicitly represents another thought/utterance.”23 It functions as a “mention” marker, a signal of a proposition implicit within the immediate discourse.24 It can also be a focus particle, used to “identify or point to a stretch of language for the application of focus.” 25 It is much like the English “that.”
Koehler and Baumgartner put the particle ki into two categories, either as demonstrative or conjunctive. As a demonstrative particle, two of its subcategories pertain to our discussion. The first is as a deictic, emphatic or stressing particle, translated as “yea, verily or indeed.” The second use finds it preceding a negation suggested by the context and translated as “Yet” in the NIV in Psalm 44:22.26 “Yet for your sake we face death all day long; we are considered as sheep to be slaughtered.”
Here we notice how important are the words “suggested by the context,” for this particle has many possibilities among others, such as adverbial (“truly, indeed”), temporal (“when”), causal (“because, for”) and demonstrative “that.” Follingstad says that, as a particle pointing to something,כי depends on its context to supply what is being pointed out more so than any other logical conjunction, such as למען or יען for example.27
The particle ki can also fulfill one more role and that is to introduce direct speech. Joüon gives the following examples of this usage where it is much like לאמר– Judges 6:16, Genesis 29:33, and Exodus 4:25.28 Others disagree on this point, saying that ki is merely an emphatic part of the direct speech and not a precursor to it.29 Follingstad also calls this use deictic or emphatic, but adds a twist, echoing the ki recitativum of Joüon and others. It “introduces new voices/speakers into the discourse or functions as a transition between discourse levels.”30 It can “initiate a domain when it occurs at the boundary between narrative and direct speech.”31
The use of γαρ in the Greek New Testament appears to be analogous to our discussion of ki in the Old Testament. Starting with an historical perspective, Robertson helps to dispel any notion we might have of Semitic influence. The New Testament use of γαρ “is in accord with that of the classic [Greek] period.” Going on, he argues that γαρ does not strictly set forth the precise relationship between clauses or sentences.32 Causal clauses are typically dependent or subordinate in nature. Nonetheless, Robertson points out, paratactic or coordinating clauses are quite common in Greek and may be introduced by γαρ and οτι.33 To these statements, Blass and Debrunner agree and add that English must often leave γαρ untranslated.34
Some of the older grammarians foresaw these issues of modern discourse theory. Thayer, for instance, stated that when γαρ is repeated in successive statements, it may confirm the same thought by as many arguments as there are repetitions of the particle, or it may introduce succeeding statements that are subordinate to one another. Sometimes γαρ confirms an entire discussion.35 This reminds us of its repeated use within the single chapter of Romans 8. The 19th century Greek grammarian, Bagster, took the particle γαρ to be mainly causal, frequently used with an ellipsis of the preceding context. However, its force could be variously represented as “now, then, to wit” (Matthew 1:18), thus introducing circumstantial details.36 Liddell and Scott held that the particle γαρ could reference “a portion only of the preceding statement, or to something implied but not expressed, [rather] than to the clause as it stands.”37
7. The nature of markers and how they can help us
Linguists have been exploring the field of discourse for the last several decades. It is still not widely respected by sentence-only grammarians. Advocates of discourse theory in the past often saw discourse markers as words or phrases, relatively syntax-independent, having no particular grammatical function within the sentence itself. In this view, these markers did not change the meaning of the utterance, and had a somewhat empty meaning. Examples of this include the particles “oh”, “well”, “now”, “then”, “you know”, and “I mean”, as well as the connectives “so”, “because”, “and”, “but”, and “or.”
An acquaintance of mine was writing his master’s thesis in linguistics some years back in French for the University of Sherbrooke in the Canadian province of Quebec. His topic was on the French word “bien,” for which he had counted over twenty different uses, many fitting quite well within the notion of a discourse marker.
As stated, in the past some of the words or phrases that were considered discourse markers were treated as “fillers” or “expletives” – words or phrases that had no function at all. Now discourse theorists are assigning them functions that can be classified into three broad groups: (a) relationships among utterances, (b) relationships between the speaker and the message, and (c) relationships between speaker and hearer.38 An example of the latter is the Yiddish nu, which acts as a marker to encourage further development within a topic.39
Discourse markers are also called connectives. Examples of such are “therefore” and “in other words” which link parts of a discourse and show their relationship. “Moreover” indicates that the upcoming text adds more information, “however” contrasts with previous information, “for instance” provides clarification or illustration, and “as a result” has a cause-and-effect relation.40
In contrast to sentence-only grammars, discourse theorists focus on larger units. The marker “and,” for instance, is a discourse coordinator which is the speaker's indicator that a discourse unit is coming up, connected with a prior discourse of an equivalent structural value. “It is a marker of speaker continuation.”41 Discourse interaction is often bracketed by “and.”42 The particle “but” is a discourse coordinator and a marker of an upcoming unit of contrasting action.43 In summary, these markers act as means for coherence and information management.44
8. Some translation examples
Test Cases: Genesis 4:23, 25; 7:4; 13:15; 37:45
KJV and NKJV translate ki as “For” where it introduces direct speech in the above verses, where NIV and NLT leave ki untranslated.
1. NKJV Genesis 4:23 Then Lamech said to his wives:
“Adah and Zillah, hear my voice;
Wives of Lamech, listen to my speech!
For (ki) I have killed a man for wounding me,
Even a young man for hurting me.”
2. NIV Lamech said to his wives,
“Adah and Zillah, listen to me;
wives of Lamech, hear my words.
I have killed a man for wounding me,
a young man for injuring me.
Test Cases: Job 3:13, 5:18, 7:21, 15:34, 20:19, 21:21, 27:8, 31:11, 32:18
The following versions all start the above verses with “For” in the sense of a causal connection dependent on the previous verse: KJV, NKJV, NIV, NLT.
Out of these nine verses in Job, the CEV begins only one verse with the causal sense of ki as is printed in the Hebrew. The way that it is translated, we see that the clausal clause is dependent not on its own main clause, but on information provided from the previous verse. The MSG, on the other hand, finds a way to eliminate the “For” and get the sentence construction correct.
1. CEV Job 21:20-21 “Let God All-Powerful force them to drink
their own destruction from the cup of his anger.
Because (ki) after they are dead,
they won't care what happens to their children.”
2. MSG “They deserve to experience the effects of their evil,
feel the full force of God's wrath firsthand.
What do they care what happens to their families
after they're safely tucked away in the grave?
Test case: John 3:16-17
The following versions begin the two verses with “For” just as it appears literally in the Greek: KJV, NKJV, RSV, WEY, YLT, NASV, NIV, TNIV, ESV, NET, GNT, AMPL.
The following versions begin verse 16 with “For,” but not verse 17: NLT, BBE, NRSV, Phillips.
The following versions start neither verse with “For”: MSG, CEV.
Examples where the verses are not started with “For”:
1. MSG : “This is how much God loved the world: He gave his Son, his one and only Son. And this is why: so that no one need be destroyed; by believing in him, anyone can have a whole and lasting life. God didn't go to all the trouble of sending his Son merely to point an accusing finger, telling the world how bad it was. He came to help, to put the world right again. Anyone who trusts in him is acquitted; anyone who refuses to trust him has long since been under the death sentence without knowing it.”
2. CEV: “God loved the people of this world so much that he gave his only Son, so that everyone who has faith in him will have eternal life and never really die. God did not send his Son into the world to condemn its people. He sent him to save them!”
Test Case: Romans 8
The following versions begin sentences with “For” in Romans 8 in the sense of a causal dependent connection with previous verses: ESV with 16 occurrences, KJV, NKJV, NASV with 14, NIV, TNIV, NLT with 7, Phillips with 4, and CEV and MSG with no occurrences. I counted fourteen sentences in the Greek text of Romans 8 that began with the Greek word γαρ. There were no textual variants for these occurrences. We are left with a choice in translation. We can write a number of dependent clauses that do not stand alone as true sentences. The alternative is to have a string of sentences, showing implicit dependency on one another due to the underlying repetition of γαρ in the Greek. The target language could either suppress this connector or explicitly represent it by using “for” or some other causal expression, but would do so this time in non-initial sentence position.
9. Concluding remarks
It would be interesting to track the use of “And, But, and For,” in English literature before the advent of the first English Bible. Would we find a lower use of these connectors in sentence initial position? How much has a word-for-word translation of the Bible impacted the current and frequent use of these connectors at the beginning of sentences? The present author remarks that certain writers never use “And” or “But” to initiate sentences. Others cannot write a page without starting a sentence with one of them. However, it is very difficult to find any modern writer who begins a sentence with “For” in the sense of a causal marker or conjunction. It would most likely occur in direct discourse, especially in response to a question. Apart from that, one can assume its practice is quite limited.
We argue here that the usage of “For” in sentence initial position is unnecessary for modern English translations of the Bible. This is in agreement with a number of well-known English dictionaries and grammars as well as Greek and Hebrew grammars of the Bible. It also follows with the research and findings of modern discourse theory linguists. Any translator of the Bible must be sensitive to context and to the way words are normally used in a target language. In addition, greater sensitivity is needed in starting sentences, paragraphs, chapters, and even books with the connectors “And” or “But.”
1 D. J. Allerton, Essentials of Grammatical Theory, London: Routledge, 1979, 203.
2 Otto Jespersen, The Philosophy of Grammar, Chicago: University of Chicago Press, 1924, 307.
3 Shorter Oxford English Dictionary on Historical Principles, Vol. 1: A-M, London and New York: Oxford University Press, 1973, Fifth Edition, 77; Philip Babcock Gove, Ph.D., Webster’s Third New International Dictionary of the English Language—Unabridged, Springfield, MA: G. & C. Merriam Co., 1961, 80.
4 Shorter Oxford English Dictionary, 314.
5 Webster’s, 303.
6 Vern Sheridan Poythress, “The Presence of God Qualifying Our Notions of Grammatical-Historical Interpretation: Genesis 3:15 as a Test Case,” JETS 50:1 (2007), 87-104.
7 Allan Harman, “Particles,” New International Dictionary of Old Testament Theology & Exegesis, ed. by Willem A. VanGemeren, Grand Rapids, MI: Zondervan Publishing House, IV, 1997, 1037.
8 Ibid. 1037-38.
9 Johannes P. Louw and Eugene Nida, eds., Greek-English Lexicon of the New Testament based on Semantic Domains, Second Edition, New York: United Bible Societies, Vol. 1, 811.
10 A. T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research, Nashville, TN: Broadman Press, 1934, 443.
11 Steven E. Runge, Discourse Grammar of the Greek New Testament: A Practical Introduction For Teaching and Exegesis, Lexham: Bible Reference Series, 18, (accessed Feb. 22, 2010), http://ntdiscourse.org/docs/Discourse grammar sample.pdf.
12 Ibid. 20.
13 Ibid. 24.
14 Wilhelm Gesenius, Gesenius' Hebrew Grammar, as edited and enlarged by the late E. Kautzsch, translated into English by Arthur Ernest Cowley, London: Oxford University Press, 2nd English edition, 1910, § 158a, p. 492.
15 Daniel B. Wallace, Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament, Grand Rapids, MI: Zondervan Publishing House, 1996, 669.
16 Ibid. 761.
17 Harman, “Particles,” 1030.
18 Ibid. 1032.
19 A. Schoors, “The Particle כי,” Remembering All the Way... A Collection of Old Testament Studies Published on the Occasion of the Fortieth Anniversary of the Oudtestamentisch Werkgezelschap n Nederland [Oudtestamentliche Studien, Deel XXI], compiled by B. Albrektson et al., Leiden: E.J. Brill 1981, 242.
20 P. Paul Joüon, Grammaire de l’Hébreu biblique, Rome: Institut Biblique Pontifical, 1987, § 164b, p. 503.
21 Bruce K. Waltke and M. O’Connor, An Introduction to Biblical Hebrew Syntax, Winona Lake, Indiana, 1990, § 39.3.1d, p. 657.
22 Ibid. § 39.3.4e, p. 665.
23 Carl M. Follingstad, Deictic Viewpoint in Biblical Hebrew Text: A Syntagmatic and Paradigmatic Analysis of the Particle כי, Dallas, TX: SIL International, 2001, 303.
24 Ibid. 151.
25 Ibid. 156.
26 Ludwig Koehler and Walter Baumgartner, revised by Walter Baumgartner and Johann Jakob Stam, translated by M. E. J. Richardson, The Hebrew and Aramaic Lexicon of the Old Testament: The New Koehler-Baumgartner in English, 5 volumes, Leiden, New York, Köln: E. J. Brill, 1994-2000, CD-Rom Edition.
27 Follingstad, Deictic Viewpoint, 141.
28 Joüon, Grammaire, 480.
29 A. Schoors, “The Particleכי ,” Remembering All the Way, 258.
30 Follingstad, 306.
31 Ibid. 308.
32 Robertson, Grammar, 1190-91.
33 Ibid. 962.
34 F. Blass and A. Debrunner, A Greek grammar of the New Testament and other early Christian Literature, Chicago: The University of Chicago Press, 1961, 235.
35 Joseph Henry Thayer, A Greek-English Lexicon of the New Testament, Edinburgh: T & T Clark, 1961, 110.
36 Bagster’s Analytical Greek Lexicon, London: Samuel Bagster & Sons, 1870, 75.
37 Henry George Liddell and Robert Scott, A Greek-English Lexicon, New York: Harper and Brothers, Publishers, 1878, 308.
39 Ghil’ad, Zuckermann, “Hybridity vs. Revivability: Multiple Causation, Forms and Patterns,” Journal of Language Contact, Varia 2, 50, (accessed Feb. 3, 2010), http://www.zuckermann.org/pdf/Hybridity_versus_Revivability.pdf.
41 Deborah Schiffrin, Discourse Markers, New York: Cambridge University Press, 1987, 141.
43 Schiffrin, Discourse Markers, 152.
44 Ibid. 74.