THE MIDDLE VOICE IN
THE NEW TESTAMENT
George J. Cline
Submitted in partial fulfillment of requirements
for the degree of Master of Theology in
Grace Theological Seminary
Digitized by Ted
Title: THE SIGNIFICANCE OF THE MIDDLE VOICE IN THE NT
Author: George J. Cline
Degree: Master of Theology
Date: May, 1983
Advisers: John Sproule; George Zemek
The middle voice in Greek has no exact parallel in the English
language. Scholars disagree about both its essential significance and
its various usages as dictated per context. The notion of voice inter-
change, i.e., usage of a middle voice with an active meaning apart
from the issue of deponency, is the primary controversy. Translational
and interpretive problems apart from voice interchange are treated as
secondary. Historical argumentation, clarification of the notion of
voice in general, and a removal of misconceptions regarding the names
of the voices are the foundation upon which ensuing argumentation rests.
The historical development of the middle voice as well as usage
invalidate the concept that the middle voice is middle in meaning between
the active and passive voices. The middle voice is older than the pas-
sive and has fluctuated in meaning with significant passage of time.
Regarding meaning of the middle voice, the suggestions of transitiveness
and general reflexivity are deemed as inadequate or misleading. Although
the concepts of special advantage and subject participation in the
results may at times be involved, these ideas are not inherent to the
middle itself. In fact, an examination of the true middles in the NT
fails to reveal a prescriptive definition applicable to every occurrence.
Instead, a basic notion of the middle voice as an intensification in
some manner or degree of the relationship between the subject and the
action expressed by the verb serves as a valid general guideline. The
absence or presence, degree, and manner of this intensification is deter-
mined by the historical development of the verb, the verbal idea itself,
and the particular context.
Voice interchange without semantic distinction is an infrequent
phenomenon in the NT. An examination of parallel synoptic passages
reveals that Mark apparently employs the middle in certain cases simply
as a stylistic variation. However, no broad spectrum principle is
available, for in James 4:2, 3 a semantic distinction is recognized,
whereas in 1 John 5:14, 15 none is apparent. Each particular case of
voice interchange should be evaluated on its own merits. In addition,
a taxonomical approach is ultimately unsatisfactory.
Several warnings are appropriate regarding the middle voice.
First, not every nuance of the middle can be expressed by English trans-
lation. Second, usage apparently varied among different authors and in
different localities. Finally, unwarranted dogmatism and insistence on
classical distinctions should be avoided. Instead, a safe guideline is
to interpret the intensification of each true middle in terms of its
context, verbal idea, and historical development.
Accepted by the Faculty of Grace Theological Seminary
in partial fulfillment of requirements for the degree
Master of Theology
John A Sproule
George J. Zemek
LIST OF ABBREVIATIONS
AJP American Journal of Philology
BAGD Bauer, W. F. Arndt, and F. W. Gingrich, Greek-English Lexicon of
the NT, rev. F. Danker
BG M. Zerwick, Biblical Greek
BGHG R. W. Funk, A Beginning-Intermediate Grammar of Hellenistic
DNTT C. Brown, Dictionary of New Testament Theology
GASS J. Thompson, A Greek Grammar, Accidence and Syntax for Schools
GLHR A. T. Robertson, A Grammar of the Greek New Testament in the
Light of Historical Research
GNTG W. F. Howard, J. H. Moulton, and N. Turner, A Grammar of New
GOECL F. Blass and A. Debrunner, A Greek Grammar of the New
Testament and Other Early Christian Literature, rev. R. Funk
HGG A. Jannaris, An Historical Greek Grammar
ICC International Critical Commentary
IJAL International Journal of American Linguistics
LPGL G. W. H. Lampe, Patristic Greek Lexicon
LSJ H. Liddell, R. Scott, and H. Jones, A Greek English Lexicon
NICNT New International Commentary on the New Testament
NTG E. Jay, New Testament Greek, An Introductory Grammar
MGNT H. Dana and J. Mantey, A Manual Grammar of the Greek New
TDNT G. Kittel and G. Friedrich, eds. Theological Dictionary of the
TABLE OF CONTENTS
List of Abbreviations vii
I. BACKGROUND 3
Meaning of Voice 3
In the active voice 6
In the middle voice 7
In the passive voice 8
History of the Voices 8
Middle Older than Passive 9
Fluctuation in Meaning 9
Names of the Voices 10
II. SIGNIFICANCE 13
Middle in Meaning 16
Special Advantage 18
Participating in the Results 18
Transitive - Intransitive 19
Fundamental Concept 21
History of the Verb 22
Idiomatic expressions 22
Distinct semantic shift 24
Form and Tense 24
III. USAGE 28
Middle for Active 30
James 4:2,3 30
Semantic difference 30
Semantic indistinction 32
1 John 5:14,15 33
Parallel Synoptic Passages 35
Matthew 26:23; Mark 14:20 35
Matthew 19:20; Mark 10:20 36
Matthew. 26:51; Mark 14:47 37
Paired Sentences 38
Using eu[ri<skw 39
Using u[stere<w 39
Using Additional Verbs 40
Active for Middle 40
Based on Similarity of Meaning 40
Based on Classical Precedent 41
Based on Different Construction 43
Passive as Middle 44
Direct Middle 47
Causative or Permissive Middle 47
Dynamic or Deponent 49
IV. TRANSLATION AND INTERPRETATION 53
Rigid Rules 54
Unwarranted Dogmatism 54
Authorial and Geographical Variation 55
Insistence on Classical Distinctions 55
For translation 56
For interpretation 57
V. CONCLUSION 58
Any thorough attempt to interpret and translate Romans 3:9
causes the exegete to ponder over the voice of proexo<meqa. Is the verb
middle or passive, or is it middle in form yet active in meaning though
not deponent? Similarly, the aorist middle participle a]pekdusa<menoj
deponent or is it a true middle with the sense of having divested himself
of something.1 The resultant theological significance is considerably
affected by the sense which is selected.2
As in the above cases, numerous exegetical questions partially
hinge upon the voice of the verb. In the case of the middle voice, the
difficulty is increased since that phenomenon is a refinement of the
Greek language that has no parallel in English. In common with other
languages of Indo-European origin, Greek expresses by inflection what
some modern languages, notably English, express by auxiliaries. Further-
more, grammarians differ in their understanding of the essential
significance of the middle voice. Thus, in order to remove some of
these obstacles, three basic problems are dealt with.
The first difficult problem concerns the elucidation of a basic
concept regarding the middle voice. After an analysis of various
1 BAGD, p. 83. They list a]pekdu<omai as deponent.
2 Homer A. Kent, Jr., Treasures of Wisdom, Studies in Colossians
and Philemon (Grand Rapids: Baker Book House, 1978), pp. 88-89. If the
verb is not deponent, then it does not properly describe the taking of
power away from evil angels.
viewpoints, a functional definition describing a basic concept of the
middle is set forth. Second, and perhaps the most controversial, are
the problematic areas of usage. Is the middle voice used with an active
meaning even though the verb is not deponent? More generally, is the
semantical distinction among the voices blurred in the NT? In addition,
the effectiveness of taxonomical approaches to usage are questioned.
Third, what are general guidelines regarding translation and interpreta-
tion of the middle voice?
Historical argumentation concerning development of the voices
combined with a clarification of the meaning of voice in general lays
the foundation for treating these problems.
BACKGROUND OF THE MIDDLE VOICE
In order to avoid semantic confusion, it is advantageous to
clarify the meaning and concept of voice as it applies to language in
general. For often the voices are treated categorically, without the
basic notion of voice having been first clarified. Also, a brief history
of the voices in Greek combined with a discussion of the terminology
relating to the voices is the necessary background for the elimination
of certain erroneous conceptions.1
Meaning of Voice
The grammatical category of voice as used by linguists and
grammarians to comprehend and analyze a specific verbal feature con-
tained in some languages has enjoyed considerable popularity over the
last few years.2 It is thus not surprising that voice as a grammatical
category has been variously defined.3 Yet, if a descriptive definition
1 Certain older grammarians are imbued with the notion that the
middle voice has a middle signification between the active and passive
voices. See, for example, Richard Valpy, The Elements of Greek Grammar
(New York: W. E. Dean, 1837), p. 82; Charles Anthon, A Grammar of the
Greek Language (New York: Harper and Bros., 1855), p. 124. They appear
to follow the precedent set by Claude Lancelot, A New Method of Learning
2 vols. trans. Thomas Nugent (
2 Jan Svartvik, On Voice in the English Verb (Hague: Mouton and
advent of transformational grammatical theory.
3 Robert J. Di Pietro, Language Structures in Contrast (Rowley:
Newbury House Publishers, 1971), pp. 75-77. A uniform descriptive
of voice is to be useful in analyzing a language, it should be suffi-
ciently general so that it does not either impose semantic restrictions
or add nuances that are not inherent in a language.1 As pertaining to
Greek, many grammarians discuss the problems of voice without clarifying
the concept of voice itself or finding any single cohesive principle
for the category.2 When the notion of voice itself is clarified it is
usually defined descriptively in terms of the relationship between the
subject of a sentence and the verbal action of its predicate.3 Simply
defined, voice is the relationship between the subject of a sentence and
the action expressed by the verb.4 The various voices indicate a range
of possible relationships between subject and predicate. Yet, strictly
definition of voice applicable to all languages is difficult to obtain.
For example, see Alice Werner, Introductory Sketch of the Bantu
Languages (London: Kegan, Paul, Trench, Trubner and Co., 1919), pp.
146-55. At least eleven different derived forms of the verb have been
found which may be described as voices.
1 Archibald T. Robertson, A Grammar of the Greek New Testament in
the Light of Historical Research (Nashville: Broadman Press, 1934), pp.
31-40 (hereafter cited as GLHR). He appropriately warns that the seat
of authority in language is not the books about language, but it is the
people who use the language.
2 Frank E. B. Leddusire, "A Comparative Study of Middle Voice in
Koine Greek and Reflexive Verbs in Old Russian through Case Grammar
Description" (Ph.D. dissertation, University of Washington, 1972),
3 For an exception, see Fred W. Householder, Kostas Kazazis, and
Andreas Koutsoudas, "Reference Grammar of Literary Dhimotiki", IJAL 30
(April 1964):102. They define voice as that which refers to the direc-
tion of the action expressed by the verb. Although this directional
concept may differentiate the active and passive voices, it appears to
be inadequate for the middle.
4 Eric G. Jay, New Testament Greek, an Introductory Grammar,
(London: SPCK, 1958), p. 14 (thereafter cited as NTG); Robert W. Funk,
A Beginning-Intermediate Grammar of Hellenistic Greek, 2d corrected ed.
vol. 2 (Missoula: Scholars Press, 1973), p. 395 (hereafter cited as
BGHG). This definition does not appear to impose upon the Greek voices
meanings that they do not contain.
speaking, voice is the property of the verbal-idea rather than of the
If a definition of voice is chosen as the relationship between
the subject and the action expressed by its verb, then for the sake of
clarity and consistency, the voices should be defined in terms of that
relationship.2 The active voice represents the subject as performing
the action of the verb. The passive voice represents the subject as
acted upon, and does not act.3 However, the middle voice denotes that
the subject is in some special manner involved or interested in the
action of the verb.4 Stated slightly differently, in the middle voice
there is an intensification in some manner between the subject and the
action expressed by the verb.5 The following examples of lou<w illustrate
Greek New Testament (New York: Macmillan Co., 1955), pp. 154-55 (here-
after cited as MGNT); Johann M. Stahl, Kritischhistorische Syntax des
der classichen Zeit (
Universitatbuchhandlung, 1907), p. 42.
2 For consistency and clarity, see Herbert W. Smyth, Greek
rev. Gordon M. Messing (
1956), pp. 389-94; Basil L. Gildersleeve, Syntax of Classical Greek,
pt. 1 (New York: Charles Scribner's Sons, 1965), p. 61-70.
3 John Thompson, A Greek Grammar, Accidence and Syntax for
Schools and Colleges (New York: E. P. Dutton and Co., 1903), p. 310
(hereafter cited as GASS).
4 Gildersleeve, Greek Syntax, 1:64.
5 A list of definitions of numerous authors was compiled. These
definitions of the voices could be divided as to the central theme. It
appears that the clearest definitions consistently define the voices in
terms of the relationship of subject and action. They virtually all
agree that there is a difference between the relationship in the
active voice and that of the middle. The relationship in the middle is
the differences between active, middle and passive voice functions,
1. h[ a]delfh> e@lousen to> te<knon. The sister bathed the child.
2. h[ a]delfh> e]lou<sato. The sister bathed (herself).2
3. to> te<knon e]lou<qh u[po> th?j a]delfh?j. The child was bathed by the sister.
The difference of emphasis between voices has been termed one of
theme, salience, or focus of attention.3 Voice per se does not appear
to place an emphasis either on the subject, the verbal action, or their
relationship. The subject or verb may be emphasized by contextual
factors such as word--order, but this is not the function of voice.4
In The Active Voice
After suggesting that the prehistoric distinction between the
active and the middle voice involved an accent on the root in the active
form and on the personal ending in the middle form, James Moulton
1 Eugene Van Ness Goetchius, The Language of the New Testament,
although not found in the NT, are particularly lucid because they emplo-
the same verb in the indicative mood. However, similar examples may be
found in the NT using lou<w, but some examples are in participial form.
For example, see e@lousen in Acts 16:33 for active; leloume<noj in John
13:10 for passive; lousame<nh in 2 Peter 2:22 for middle.
2 This use of the middle as reflexive is only one of the possible
functions of the middle voice. No single example can be cited to illus-
trate the broad spectrum of possibilities.
3 Herbert H. Clark, Semantics and Comprehension (Hague: Mouton
the emphasis of actives and passives in English, see p. 118.
4 GLHR, p. 798. His statement that the use of voice is to
direct attention to the subject, not to the object, may be misleading.
It should be noted that this statement is made regarding
conjectures that originally in the active the action was stressed, in
the middle the agent.1 However, this possible historical distinction
does not appear to be the case in NT usage as illustrated by John 14:1.
pisteu<ete ei]j to>n qeo<n, kai> ei]j e]me> pisteu<ete. By means of a chiasm the
two verbs are placed in two emphatic positions, stressing the durative
action of believing.2 In the following verse ei#pon is not in an emphatic
position, and it is difficult to envision that the active voice of ei#pon
emphasizes the act of speaking. It simply indicates that Jesus, the
subject, is the performer of the action.
In The Middle Voice
Similarly, the assertion that the middle voice stresses the
agent needs to be either qualified or avoided. Dana and Mantey carefully
explain this notion with the following considerations.
While the active voice emphasizes the action, the middle stresses
the agent. It, in some way, relates the action more intimately to
the subject. Just how the action is thus related is not indicated
by the middle voice, but must be detected from the context of the
However, it appears possible to relate the action more intimately
to the subject without necessarily stressing the subject, i.e., the
agent of the action being the focus of attention rather than the rela-
tionship between the subject and the action. For example, katalamba<nw
in the active voice means to seize or overtake, but in the middle denotes
grasping for oneself or with reference to oneself, and thus to
comprehend. A mental as opposed to a physical application of katalamba<nw
1 GNTG, p. 512.
2 R. C. H. Lenski, The Interpretation of
(Minneapolis: Augsburg Publishing House, 1943), p. 969.
3 MGNT, p. 157.
is introduced by the middle in this way, since mental action is
especially confined within the sphere of the agent.1 Hence the subject
of this verb in the middle voice indicates both the performer of the
action and that to whom or for which the action is performed.2 If this
notion is justifiably considered as stress, it is certainly far less
emphatic and of a different nature than the stress of a subject as indi-
cated by a personal pronoun as in the following example. ]Egw> de>
katelabo<mhn mhde>n a@cion au]to>n qana<tou pepraxe<nai. "But when I
understood that he had committed nothing worthy of death" (Acts 25:25).3
Thus, if one wishes to speak of special attention being focused
on the subject by the middle voice, it is only in the sense that the
subject both performs the action and is that to whom or for which the
action is performed.
In The Passive Voice
Similarly, the passive voice simply represents the subject as
being acted upon. Any notion of emphasis regarding the subject, verb,
or their relationship is due to contextual factors.
History of the Voices
The question regarding the antiquity and development of the
voice forms has not been fully established, and the gaps in knowledge
are often the areas of much conjecture.4 Yet there does appear to be
1 Wilbert F. Howard, James H. Moulton, and Nigel Turner, A Grammar
of New Testament Greek, vol. 1 (Edinburgh: T. and T. Clark, 1906),
p. 158 (hereafter cited as GNTG).
2 Goetchius, Language of the New Testament, p. 104.
3 This author is responsible for the translations of Greek
statements throughout this thesis.
4 GNTG, 1:152-53.
sufficient historical information to establish that the middle is prior
to the passive in historical development.
Middle Older Than Passive
Although it is unknown whether the active or the middle voice
was the first to develop, it is generally recognized that primitive
Greek, as in other Indo-Germanic languages, had only two voice forms,
active and middle.1 The middle form was subsequently more fully devel-
oped into the passive.2 During the Attic period a complete system of
three voices existed.3 The ensuing tendency during the Hellenistic per-
iod was to merge the middle and passive forms into a single form with the
passive gaining ascendancy.4 In modern Greek, there is no middle form.5
Fluctuation in Meaning
Although John Thompson asserts that the original sense of the
middle form was reflexive, it appears that this is questionable.6 Yet
1 Karl Brugmann, A Comparative Grammar of the Indo-Germanic
Languages, vol. 4, trans. R.
Seymour Conway and W. H. D. Rouse (
B. Westerman and Co., 1895), p. 515; Satya S. Misra, A Comparative
Grammar of Sanskrit, Greek and Hittite, with a Foreward by Sunuti K.
Chatterji (Calcutta: World Press Private, 1968), p. 90.
2 James H. Moulton, An Introduction to the Study of New Testament
Greek, 5th ed., rev. Henry G. Meecham (London: Epworth Press, 1955),
p. 41. For a different viewpoint, see GASS, p. 305. Yet he still recog-
nizes middle is older than passive.
3 Anthony N. Jannaris, An Historical Greek Grammar (
Macmillan and Co., 1897), p. 362 (hereafter cited as HGG)
4 Friedrich Blass and Albert Debrunner, A Greek Grammar of the New
Testament and Other Early Christian Literature, trans. and rev. Robert W.
Funk (Chicago: University of Chicago Press, 1961), p. 161 (hereafter
cited as GOECL). For probable causes of this merger, see HCG, p. 362.
5 Irene P. Warburton, "On the Verb in Modern Greek" (Ph.D.
6 GNTG 1:156. Although a reflexive meaning ultimately accrued to
the middle form, it would be wrong to assume that it was originally
whether or not this is true for certain periods, it is not true of NT
usage.1 The voices do vary in their usage during different stages of
the language.2 Although in the NT the middle forms may still retain a
wide field of usage for all the senses found in classical use, there are
examples contrary to the general trend.3 Thus, one should not evaluate
usage of the middle voice form in the NT solely by classical standards
or consider NT writers as lacking in their understanding of certain
Names of the Voices
The names and earliest descriptions of the verbal category of
voice have been traced to Dionysius Thrax.5 Grammarians have objected
to the terminology of the Greek voices as not being clearly descriptive
of usage. Active is not distinct for the other voices also express
there. For a discussion of the controversy regarding reflexivity in
voice, see Leddusire, "Middle Voice," pp. 36-37.
1 C. F. D. Moule, An Idiom Book of New Testament Greek, 2d ed.
(Cambridge: Cambridge University Press, 1959), p. 24.
2 GLHR, p. 799.
3 Maximilian Zerwick, Biblical Greek, adapted from the 4th Latin
ed. by Joseph Smith (Rome: Pontificii Instititi Biblica, 1963), pp.
75-76 (hereafter cited as BG).
4 GLHR, p. 805;
Edwin Hatch, Essays in Biblical Greek (
At the Clarendon Press, 1889), pp. 2-8.
5 Dionysius Thrax, Grammatici Graeci, vol. 1 (Lipsiae: In
Aedibus B. G. Teubneri,
1838; reprint ed.,
Verlagsbuchhandlung Hildescheinz, 1965), pp. 48-49. His term for voice,
diaqe<sij, includes the three terms ene<rgeia meso<thj and pa<qoj. For
further history of the terminology, see F. E. Thompson, A Syntax of
Attic Creek (London: Longmans, Green and Co., 1907), pp. 158-59; Basil
L. Gildersleeve, "Stahl's Syntax of the Greek Verb," American Journal
of Philology, 29 (1908):275.
action.1 Furthermore, the active does not always express an action,
but may denote a state.2 Concerning the middle, it does not stand in
between the active and passive in meaning.3 But even more objections
are raised against the name of deponent.4 This term is derived from the
Latin depono meaning to lay aside, since these verbs appear to have laid
aside and lost the active form.5 Yet certain verbs are found in the
active form only or the middle form only, and thus Moulton would prefer
to apply the name of deponent to both of these classes.6 Although it
may be recognized that the terms are not clearly descriptive of usage,
the solution does not appear to be the coining of new terms in place of
those which are imbedded in grammars and history. Instead, these terms
should be properly defined in terms of their usage.
1 GLHR, p. 331.
2 Friedrich Blass, Grammar of New Testament Greek, 2d ed. rev.
and enl., trans. Henry Thackeray (London: Macmillan and Co., 1905), pp.
180-81. However, linking verbs are best understood apart from the active
or passive idea. For example, see BGHG, 2:398-99.
3 GLHR, p. 331.
4 Certain grammarians even attempt to make deponents a different
category from middles. For example, see George B. Winer, A Grammar
Idiom of the New Testament, 7th ed. enl. and imp. Gottlieb Lunemann
(Andover: Warren F. Draper, 1869), p. 258. He proposes that from
middle verbs are to be carefully distinguished deponents. To eliminate
the confusion regarding deponents, sometimes a non-deponent is called a
true middle. For example, see BGHG 2:398. Others use the term
defective rather than deponent.
5 NTG, p. 85. But in some cases these verbs never had an active
form. A deponent is more accurately define as a verb which has an
active meaning, but only middle (or middle and passive) forms.
6 GNTG, 1:153.
The grammatical category of voice indicates how the subject is
related to the action expressed by the verb. The active voice repre-
sents the subject as performing the action of the verb. It simply
represents the subject as acting without necessarily stressing the
action. The passive voice simply represents the subject as being acted
upon. The middle voice indicates an intensification in some manner
between the subject and the action expressed by the verb, i.e., the
subject is in some special manner involved or interested in the action
of the verb. Although certain grammarians assume that the middle voice
stresses the agent of the action, this is valid only in the sense that
the subject both performs the action and is that to whom or for which
the action is performed. An examination of the history of the voices
invalidates the erroneous concept that the middle voice is middle in
meaning between the active and passive, for the middle form is older
than the passive form. Also from the historical survey it is seen that
the voices have varied in their usage during different stages of the
language. Thus classical standards, by themselves, are not a proper
criterion for evaluating NT usage. Finally, it is recognized that the
names of the voices are not clearly descriptive of their function, and
one should not be misled by the names. Instead, the terms should be
properly defined as regarding their usage.
SIGNIFICANCE OF THE MIDDLE VOICE
Up to this point it has been briefly assumed, but not proven,
that the middle voice denotes that the subject is in some special manner
involved or interested in the action of the verb. Stated slightly dif-
ferently, in the middle voice there is an intensification in some manner
or degree between the subject and the action expressed by the verb.1
However, this assumption needs to be both clarified as well as qualified.
For it is correctly maintained that it is scarcely possible to formulate
a single definition of its basal function which could be applied to all
its actual occurrences.2 For such a definition, when applied to
particular cases, is subject to limitation or even contradiction.3 An
inductive approach to the study of true middles appears to confirm this,
for no single principle has been found which captures the meaning of
every true middle.4 Moulton even asserts that it is useless to exercise
1 For the difficulty involved in selecting a theoretical frame-
work for the study of voice problems, see Leddusire, "Middle Voice,"
p. 8. He rejects the traditional descriptive approach and adopts gener-
ative transformational grammar in the tradition of Noam Chomsky as the
only adequate basis. However, traditional grammar, which defines parts
of speech by their meaning and function, is fully capable of providing a
functional basis for the formulation of a workable definition.
2 MGNT, p. 157.
3 Blass, Grammar of New Testament Greek, p. 186.
4 A printout of all the middles in the NT was obtained from
project GRAMCORD. The printout of the middles was in two separate lists,
being separated on the basis of deponency. The majority of the
middles in the NT are deponent.
one's ingenuity in interpreting every middle, for the development in
some cases never progressed beyond the rudimentary stage.1
Thus, this assumption of intensification by the middle will be
first clarified and qualified by surveying different viewpoints among
grammarians. Second, examples and data that do not fall under this
general guideline will be examined.
Although some grammars do have a general functional definition
of the middle voice, the following viewpoints of mediality are either
inadequate, misleading, or too vague to provide a clear operational
The term "reflexive," as found among different grammarians, was
rarely limited to a directly reflexive sense, i.e., the action is
directly referred back to the subject. The notions of reciprocity,
indirectness, and self-interest are sometimes included.2 Because of this
broad semantic extension, this is a difficult concept to analyze as
regarding its involvement in any basic notion of mediality.
Jelf clearly maintains the reflexive position.
The essential sense which runs throughout the middle reflexive
verb is Self--the action of the verb has immediate reference to
self. This is the proper generic notion of all middle verbs, and
1 GNTG 1:158. His statement regards the category of dynamic mid-
dles. Yet this does not mean that a general function does not belong to
the middle voice. Usage over time may fix a different idiomatic meaning
to a middle, and thus it does not reflect the general function.
2 HGG, p. 360.
the particular sense of each middle verb must be-determined by dis-
covering the relation in which that notion of self stands to the
notion of the verb.1
Curtius and Sonnenschein also maintain that the basic notion of
the middle is primarily, but not exclusively, reflexive.2 Evidence for
this position is not lacking among the middles of the NT.3
Jay denies a reflexive usage of the middle in the NT in the
direct sense. "The beginner is apt to jump to the conclusion that the
Greek Middle Voice is reflexive. This is not so. It denotes that the
subject performs the action for himself, but not to himself."4 However,
the following two examples of directly reflexive usage invalidate his
1 William E, Jelf, A Grammar of the Greek Language, 2d ed. 2
vols. (Oxford: James Wright, 1851), p. 14. Yet he maintains that
reflexivity is distinct from reciprocity and divides middles into two
categories: reflexive and reciprocal. For a similar position, see
Raphael Kuhner, Grammar of the Greek Language, for the Use of High
Schools and Colleges, trans. Bela B. Edwards and Samuel H. Taylor
(Andover: Allen, Morrill and Wardwell, 1844), p. 330.
2 Georg Curtius, The Greek Verb: Its Structure and Development,
Augustus S. Wilkins and Edwin B. England (
1880), p. 55. He uses the term "reflexive" in the broadest sense of the
term, not simply the direct passing of the action back onto the subject.
see Basil F. C. Atkinson, The Greek Language (
1931), p. 136; Edward A. Sonnenschein, A Greek
Kegan, Paul, Trench, Trubner and Co., 1914), p. 274.
3 For specific examples see pp. 47-48.
4 NTG, p. 14.
5 For a different view of a]ph<cato, see CNTG 1:155; Moule, An
Idiom Book of the New Testament Greek, p. 24. But the suggestion of the
English intransitive choke is not warranted by the details of the
parallel account in Acts 1:18. Secondly, it has been observed that the
only middle for self-murder is a]ph<cato which seems to have been the most
natural form of self-murder. Gildersleeve, Syntax of Classical Greek,
1. kai> a]pelqw>n a]ph<gcato "And after he departed, he hung himself"
2. o!ti yu?xoj h#n, kai> e]qermai<nonto h@n de> kai> o[ Pe<troj met ] au]tw?n e[stw?j
qermaino<menoj "Because it was cold and they were warming themselves;
And Peter also was with them standing and warming himself" (John
As regarding reflexivity in the broader sense, Leddusire has
concluded that although Koine mediality can include underlying reflexive
constructions, the notion of reflexivity should not be considered the
primary motivation for voice.1
Although Robertson observes that reflexive is a better.designa-
tion of the middle than the tern: "middle" if direct reflexive is not
meant, the reflexive notion does not appear to be sufficient in relating
a basic concept regarding the middle voice for several reasons.2 The
sense of indirect reflexivity is very vague and differs from author to
author.3 It is unclear as regarding its termination point, for when
does a middle cease to be indirectly reflexive. Second, it is very
imprecise regarding the function of voice. The notion of emphasis,
either subject, verbal-action, or an interaction, is not specified.4
Middle in Meaning
The position maintained by Anthon, Valpy, and Lancelot that the
middle voice form is middle in meaning is modified by Wenham.
1 Leddusire, "Middle Voice," p. 56.
2 Ibid., p. 331.
3 For example, see Gildersleeve, Greek Syntax 1:64. In some of
its uses, the middle corresponds to the English reflexive, but the
signification is much wider and shades off from what is practically a
direct reflexive until it ceases to present any translatable difference
from the active.
4 For discussion of this problem see the section on emphasis, p. 6.
Though some forms of the Middle are the same as the Passive, the
Middle is in meaning much closer to the Active than the Passive.
In fact, the meaning of Active and Middle are often indistinguish-
able. It is better to think of the Middle as a sort-of-Active than
as a sort-of-Passive.1
This modification, although not as directly erroneous as Anthon's
position, is still inadequate. Sometimes the middle may appear to be
closer to a passive idea than an active notion.2 Common ground between
the middle and passive is to be observed in the examples of which a
translation submit to or let oneself be is often suggested for the middle.
For example, a]dikei?sqe is present middle or passive in form (1 Cor 6:7).
BAGD, apparently taking this verb as a middle, offers the translation
let oneself be wronged.3 Zerwick understands this verb to be passive
and translates suffer an injustice.4 The context appears to place the
responsibility on the subject of a]dikei?sqe, and hence the middle is
appropriate. They ought to have submitted to injustice, to have ignored
their rights, to have allowed themselves to be defrauded.5 In this case,
the subject not only performs an action, i.e., letting or permitting
oneself, but also by implication is acted upon, i.e., is wronged.
Although this is not the same as the passive be wronged in every case,
1 John H. Wenham, The
Elements of New Testament Greek (
Valpy, and Lancelot, which were discounted via historical argumentation,
see p. 3.
2 GNTG, 1:162.
3 BAGD, p. 17. The verb, when taken as passive, is translated as
be wronged, be unjustly treated (Acts 7:24; 1 Cor 6:7).
4 Mary Grosvenor and Max Zerwick, A Grammatical Analysis of the
Greek New Testament, vol. 2 (Rome: Biblical Institute Press, 1979),
5 James L. Boyer, For A World Like Ours, Studies in 1 Corinthians
(Winona Lake, BMH Books, 1971), p. 70.
for one can be wronged by force without being a cooperative participant,
Moulton correctly notes that the dividing line between middle and passive
in such cases is a fine one at best.1
The attempt to precisely describe and define the relationship of
the subject to the verbal-action in the middle voice may lead one into
error. Although the agent of the action may be stressed, this does not
mean that the action described is necessarily of special advantage or
significance to the subject as proposed by Jay.2 He hung himself,
a]ph<cato, was certainly not of special advantage or significance to Judas
Similarly, it is difficult to envision that special advantage or
significance for the subject is being emphasized by ai]wni<an lu<trwsin
eu[ra<menoj "having obtained eternal redemption" (Heb 9:12). Instead, he
found the way. Jesus is represented as having secured eternal
redemption by himself.3
Participating in the Results
Dana and Mantey comment that the middle voice is that use of
the verb which describes the subject as participating in the results of
the action.4 However, they carefully expand this concept by adding that
the middle, in some way, relates the action more intimately to the
1 GNTG 1:162. Also perite<mnhsqe in Gal 5:2.
2 NTG, p. 14.
3 GLHR, p. 809. For a different rendering of this middle see
James Moffatt, A Critical and Exegetical Commentary on the Epistle to
the Hebrews, ICC (Edinburgh: T. and T. Clark, 1924), p. 121.
4 MGNT, p. 157.
subject. The precise manner in which the action is thus related to the
subject is not indicated by the middle voice.1 Similarly, Gideon and
to the subject as in some way participating in the results of the ac-
tion.2 Subject participation is clearly not always the case, since the
middle may represent the agent as voluntarily yielding himself to the
results of the action, or seeking to secure the results of the action in
his own interest.3 For example, the woman does not appear to be parti-
cipating in the results of the command keira<sqw, "For if a woman will
not wear a veil, let her also have her hair cut off" (1 Cor 11:6).
Thus, while subject participating in the results may at times
be involved, this is not a fundamental concept regarding the middle.4
Transitive - Intransitive
Transitivity has been associated with voice as early as Jelf.5
The issue of transitivity obscures the notion of voice, and makes the
discovery of any general notion of voice more difficult.6 To state the
difference between active and middle as merely that of transitive and
1 Ibid., p. 157.
2 Virtus E. Gideon and Curtis Vaughan, A Greek Grammar of the New
Testament (Nashville: Broadman Press, 1979), pp. 91-92. More generally
it (the middle voice) represents the subject as acting in relation to
himself--either on himself, for himself, or by himself.
3 MGNT, p. 160. The example in 1 Cor 11:6 was chosen because the
foam is middle aorist imperative, and thus the problem in other examples
concerning identity of a middle-passive form is avoided.
4 Even this particular nuance is not an inherent feature of the
middle. The precise relationship of the subject with reference to him-
self is not indicated by the middle itself.
5 Jelf, A Grammar of the Greek Language, pp. 10-15.
6 Leddusire, "Middle Voice," pp. 26-30. His analysis of this
problem is particularly lucid.
intransitive is incorrect.1 Voice per se does not deal with the ques-
tion of transitive or intransitive action.2 Robertson rejected transi-
tivity as being essential to voice.3 His forceful argument consists of
four observations. First, any one or all of the voice forms may be in
association with transitive verbs. Second, an inherently intransitive
verb like gi<nomai can appear in any voice form without its intransitivity
being lost.4 Third, a verb may be both transitive and intransitive in
the same voice. Fourth, transitivity varies in different languages
because it relates to the restrictions of a particular verb.5 However,
both transitivity and voice are properties of the verb.6 But transiti-
vity is discerned by the relation of the verb to an object, and is
determined by the nature of the verbal idea. Voice, also a property of
the verbal idea, indicates how the subject is related to the action.7
A survey and analysis of selected viewpoints among grammarians
has yielded the following results. Although direct reflexivity does
1 Atkinson, The Greek Language, p. 136.
2 Smyth, Greek Grammar, p. 393.
3 GLHR, pp. 330-31. These arguments are clearly summarized and
presented with examples by Leddusire, "Middle Voice," pp. 28-30. He
adds a fifth argument that intransitive middle or reflexive verbs may
in fact represent an underlying verb with an object. This would mean
that they are only overtly intransitive, while in underlying grammar
they serve a transitive-like function.
4 e]ge<neto middle deponent, to> ge<gonoj active, genhqeh<tw passive.
5 GLHR, p. 330.
6 There are exceptions. Some verbs do vary according to form.
Thus, i!sthmi, a regularly transitive or causative verb, has an
intransitive sense in the perfect and second aorist. For discussion,
Samuel Green, Handbook of the Greek New Testament (
Fleming H. Revell Co., 1880), p. 292.
7 MGNT, pp. 154-55.
occur among NT middle forms in a few cases, the reflexive notion does
not appear to be sufficient in relating a basic concept of the middle.
The suggestion of indirect reflexivity is too general and vague, and the
usage of this term differs among various authors. Also indirect reflexi-
vity is very imprecise regarding the function of voice, for the notion
of emphasis is not specified. Subject participation in the results of
the action at times may occur as a usage of the middle, but this is not
a universal concept inherent in the middle voice itself. The precise
manner in which the action is related to the subject is not indicated by
the middle voice. Likewise, transitivity is not a concept essential to
voice. Voice does not deal with the question of transitive or intransi-
tive action. Also the middle voice is not middle in meaning between
active and passive. Nor is the suggestion that the middle voice is in
meaning much closer to the active than the passive particularly helpful,
for sometimes the middle may appear to be closer to a passive idea than
an active notion.
The suggestion, however, that the middle voice denotes the sub-
ject in some special manner involved or interested in the action of the
verb does appear to be a valid principle.2 It serves as a general
guideline when applied to true middles.3 Yet even this general notion
1 MGNT, pp. 154-55.
2 Gildersleeve, Greek Syntax, 1:64. For a brief summary of
opinions that attempt to represent a similar notion, see MGNT, p. 157.
3 Again, it is important to note the basis upon which this sugges-
tion is considered valid. Since an inductive approach to the study of
the middles of the NT has failed to reveal a basic principle that is
applicable to every middle, the best functional definition by a grammar-
ian that appears to be valid in the majority of cases was selected.
does not cover every middle, and thus needs to be qualified by the
History of the Verb
A survey of the history of a verb from its earliest traceable
origin down to the time of the usage under consideration may indicate
that there is no exegetical significance of the middle voice in terms of
this general guideline. For a historical survey of the verb may reveal
an idiomatic usage of the middle that has become established over time,
a possible deponent usage not necessarily indicated by a lexicon, or a
distinct semantic shift of meaning from active to middle.
The verb poie<w in its middle form followed by a verbal noun in
classical Greek formed a periphrasis for the simple corresponding verb.2
Although bebai<a poiei?sqai is rendered by Lenski as continuous making
sure and firm for ourselves in 2 Peter 2:2, the expression may simply
have the same sense as the verb bebaio<w.3 Another idiom listed by
Robertson is a future middle form of a verb which has a passive
meaning.4 On the basis of the future middle form being used in
1 For a more extensive treatment of these issues, see chapter
three, "Usage of the Middle Voice."
2 Smyth, Greek Grammar, p. 391. See pp. 22-23.
3 R. C. H. Lenski, The Interpretation of I and II Epistles of
the three Epistles of John, and the Epistle of Jude, (
the basis of historical precedent, they state that the middle of poie<w
serves mostly as a periphrasis of the simple verbal idea.
4 GLHR, p. 819. Considering the rather large list of verbs that
once used the middle future as passive in sense, the idiom is rare in
passive sense by Homer, Herodotus, Thucydides, Xenophen, Plato and
Demosthenes, as well as having been identified as occurring in the LXX,
he suggests the possibility that peribalei?tai in Rev 3:5 and a]poko<yontai
in Gal 5:12 may be examples of this idiom.1
Also a survey of the historical evolution of a verb may indicate
a prior history of deponent usage.2 For even if a verb occurs in both
an active form and a middle form in the same tense among literature
written within the same time period, this still may not be an indication
of a true middle, i.e., non-deponent middle. For example, in classical
Greek of the Attic period the future form of a]kou<w is regularly deponent
as a]kou<somai.3 However, in the NT the verb is usually cited as active
in its second principal part as a]kou<sw.4 The verb only occurs eight
times in the future tense in the NT with four forms being active and
four forms being middle.5 Since there is no obvious nuance intended by
1 Ibid., p. 819. For strong argumentation against this idiom in
Gal 5:12, see John Eadie, A Commentary on the Greek Text of the Epistle
Paul to the Galatians (
cited, however, to simply illustrate the importance of considering the
historical evolution of a verb as one of the factors to be considered
when evaluating the possible exegetical signifiance of a middle form.
2 See pp. 49-50 for further discussion.
3 Joint Association of
Classical Teachers Greek Course,
Grammar, Vocabulary and Exercises (
University Press, 1978), p. 284.
4 MGNT, p. 255; J. Gresham Machen, New Testament Greek for
Beginners, (Toronto: Macmillan Co., 1951), p. 255. However, both forms
for the future are listed in GNTG 2:227.
5 Alfred A. Geden and William F. Moulton, eds., A Concordance to
rev. Harold K. Moulton (
the future middle a]kou<somai in its contexts, the historical precedent of
deponency in classical Greek contributes to the decision that these
middle futures in the NT are deponent.1
Distinct semantic shift
Occasionally the middle form of a verb expresses a distinct
semantic change as compared to the active form and is best translated
as an active voice with a different meaning.2 These distinct differ-
ences, such as a@rxw (I rule) but a@rxomai (I begin) usually pose no
problem as they have well-known lexical meanings. However, similar
shifts occur for verbs that are not as well known. For example, the
imperfect a]pelu<onto in Acts 28:25 apparently simply means were going
Form and Tense
The Koine Greek verbal system consists of two forms, the finite
and the non-finite. Finite forms are sub-categorized by moods, wheres
non-finite forms are subdivided as infinitival, participial, and verbal-
adjectival in -teoj.4 Of the non-finite forms, the infinitive
1 GLHR, p. 333. He cites at least 15 verbs which had the future
in the middle form as deponent in classical Greek but have an active
future form in the NT. In the case of a]kou<w, apparently this transition
is not complete. Also note zh<sw and zh<somai in NTG, p. 319.
2 GLHR, p. 804. His attempts to trace the middle meaning of
verbs of this type to an original reflexive sense are not always
possible. For example, game<w (I marry, used of the bridegroom) but
game<omai (I marry, used of the bride). Similarly gra<fw (I enrol) but
gra<fomai (I indict).
3 BAGD, p. 96. A sufficient number of passages are cited with a
parallel meaning. Although it is not difficult to envision how this
sense could have been developed in the middle as compared to one of the
active meanings, to send away.
4 Leddusire, "Middle Voice," p. 42.
apparently did not originally possess voice functions.1 Robertson postu-
lates that gradually by analogy the infinitive forms came to be
associated with the voices in the moods.2 Gildersleeve warns against
always assuming voice significance in an infinitive.
The infinitive being a verbal noun is not so strictly bound by the
voices as the finite form. The infinitive as a complement to
adjectives and the so-called epexegetic infinitive often coincide
with the English idiom in which good to eat is good for food.3
In this regard Robertson appears correct in asserting that there
is no special voice significance in fagei?n in the phrase kai> ei#pen
doqh?nai au]th? fagei?n "and he said that something to eat be given to her"
(Mark 5:43). For the infinitive fagei?n, being a verbal-noun, serves as
the accusative of general reference of doqh?nai.4 However, his remark
that after the infinitive is fully developed its voice appears exactly
as in the moods is not particularly lucid. How does one determine in
the NT if an infinitive is "fully-developed" or in primitive form?5
Regarding voice in a participle it appears correct to understand that
all the nuances of the voices appear in the participle, and the voices
in the participle parallel usage in the finite verb itself.6
1 GNTG, 1:203.
2 GLHR, p. 1079.
3 Gildersleeve, Greek Syntax, 1:63.
4 GLHR, pp. 1079-80.
5 Few grammarians deal with this issue. But see Leddusire
"Middle Voice," p. 42. He cogently argues that the voice idea is re-
duced in infinitive forms, perhaps because of the derived nature of the
infinitive phrase, the usual deletion of the subject of the infinitive
phrase, and the absence of person indicators.
6 GLHR, p. 1110-11. This assertion is supported by the evidence
that voice appears in the earliest Greek participles as well as Sanskrit.
Also the examples cited by Robertson give ample proof of active, middle,
and passive voice distinctions in participles in the NT. Furthermore,
no participles have been encountered which do not admit a possible voice
distinction, nor has any grammarian been found to suggest otherwise.
Concerning voice in a finite form a change of mood does not
appear to cause a fluctuation in the significance of the voice.1
However, a change in tense may affect the significance of a middle form
on the basis of deponency. A verb which is not deponent in one principal
part may be deponent in another part.2
Although no single principle was discovered from an inductive
study of middles in the NT that is valid for every occurrence of a true
middle, the suggestion that the middle voice depicts the subject as in
some special manner involved or interested in the action of the verb
serves as a general guideline in the majority of cases in the NT.
However, this significance should not be automatically attributed to
every true middle. A survey of the historical evolution of a verb may
indicate idiomatic usage of the middle, possible deponent indications
which may not be lexically cited, or a distinct semantic shift that has
become fixed over a limited time period.
Also the form and tense need to be considered when evaluating
voice significance. Although all finite forms of a verb and the parti-
ciple demonstrate distinct voice functions, this is not always the case
of an infinitive, especially when used as a complement to adjectives and
in epexegetical usage. Regarding tense, it is important to know the
principal parts of a verb. For a shift from active to middle voice form
1 The monumental task of deductively studying mood shifts to
ascertain this assertion has not been done. However, again, no negating
evidence has been encountered nor has any grammarian been found to
2 This is especially true regarding future deponent middles of
many non-deponent present tense verbs. For example see the list in
with a shift in tense, such as present to future, may simply be a
transition to a deponent form.
USAGE OF THE MIDDLE VOICE
Although the middle voice signals an intensification in some
degree or manner between subject and action expressed by its verb, what
this precise intensification is, the middle voice per se does not
indicate.1 The nature of this intensification must be derived from the
context, the historical development of the verb, and the significance
of the verb itself.2 Thus, usage is the key. Gildersleeve maintains
that the interpretation of the differences between active and middle are
not so much grammatical as lexical.3 The grammatical definition does
not determine the practical use, the conventional use. Thus, gh?mai is
used of the man and gh<masqai of the woman.4 However, these differences
of interpretation are not due to features inherent in the voice itself.
When analyzing usage of the middle voice in the NT, grammarians often
center their discussions around two phenomena. First, there is the
purported usage of the middle voice which overlaps or is synonymous with
the active and passive voices. Second, there are usages in which the
middle voice expresses a distinct nuance, and these nuances are usually
treated with a taxonomical approach.
1 GNTG 1:41; William H. Davis, Beginner's Grammar of the Greek
New Testament (New York: George H. Doran Co., 1923), p. 37.
2 GLHR, p. 804.
3 Gildersleeve, "Stahl's Syntax of the Greek Verb," p. 277.
4 Ibid., p. 277.
Turner asserts that during the New Testament period there was
much confusion of meaning between the active and middle voice forms, and
the middle form was a luxury which was dispensed with in time. New
Testament authors were rapidly losing their grip on nice grammatical
distinctions in voice.1 An even more vague generalization reached by
Simcox is that although perhaps the distinction is beginning to be
blurred among some of the NT writers, it is preserved to a greater or
lesser extent in most.3 While recognizing possible overlap, Moulton
agrees with the summary of Blass that on the whole NT writers were per-
fectly capable of preserving the distinction between the active and the
middle.4 This more reserved conclusion is also arrived at by Zerwick,
who notes that on careful examination, the use of the active can usually
be accounted for.5 In view of this controversy, the specific examples
cited as support need to be evaluated. The passages pertaining to this
controversy may be aligned under three headings: middle for active,
active for middle, and passive for active or middle.6
1 Nigel Turner, Grammatical Insights into the New Testament
(Edinburgh: T. and T. Clark, 1965), p. 112.
2 Moule, Idiom Book, p. 24.
3 William H. Simcox, The Language of the New Testament, Reprint
ed. (Winona Lake: Alpha Publications, 1980), p. 95.
4 GNTC 1:158; Blass, Grammar of New Testament Greek, p. 95.
5 Zerwick, BG, p. 73.
6 Allen C.
the Gospel according to St. Matthew, ICC 3d ed., ed. C. A. Briggs,
et al. (Edinburgh: T. and T. Clark, 1912), p. xxiii. He also
uses a fourth category of active for passive.
Middle for Active
Turner, an avid proponent of the interchangeability of voice
forms without a difference in meaning, declares the following bold
While it is true that the lexicons provide no example of the middle
voice being used in an active sense, the New Testament abounds
(emphasis mine) in instances where a middle voice is used when
there is an active form of the verb available; indeed, the middle
is often used in the very sentence where its active form occurs
with the same meaning.1
However, one certainly hesitates to subscribe to such a dictum
without solid evidence.2 Indeed, the passages usually cited are few in
number, with James 4:2 being given as the classic example of voice
In this passage the same verb ai]te<w alternates in voice between
middle, active, and middle, respectively. "You do not have because you
do not ask (dia> to> mh> ai]tei?sqai u[ma?j). You ask (ai]tei?te) and do not
receive, because you ask with wrong motives (kakw?j ai]tei?sqe), so that
you may spend it on your pleasures" (Jas 4:2, 3). Numerous and varied
attempts to explain this interchange of voice in terms of a definite
semantic difference have been set forth.
Mayor suggests that a slight additional shade of meaning is
added by the middle voice. The active suggests using the words without
1 Turner, Grammatical Insights, p. 106.
2 The purported numerous passages are not cited by the author.
3 Leddusire, "Middle Voice," p. 127.
the spirit of prayer, while the middle means asking with the spirit of
prayer.1 However, the context does not support this suggestion. For
how can one ask with wrong motives (kakw?j ai]tei?sqe) with a true spirit
of prayer?2 On the other hand, to ascribe an un-prayerlike request to
the voice of ai]tei?sqe as the reason for its being kakw?j is to ignore
dia> to> mh> ai]tei?sqai which states that one does not have what he needs
because he does not ask in that very verbal voice.3
Zerwick finds the difference between middle and active to be
especially clear when the same verb is used in the same context in both
verses.4 Thus, Mark makes a quite classical distinction between ai]te<w
simply ask, and ai]tou?mai avail oneself of one's right to ask. "And he
swore to her, 'whatever you ask (ai]th<shj) of me, I will give it to you;
up to half of my kingdom.' And she went out and said to her mother,
'What shall I ask (ai]th<somai)?'" (Mark 6:23, 24).5 So also the same
distinction may be in James 4:2, 3.6 Hiebert agrees that the middle here
retains its usual middle force of to ask for your own selves since the
purpose clause in verse three certainly involves this personal interest
1 Joseph B. Mayor, The
Epistle of St. James, 3d ed. (
Macmillan and Co., 1913), pp. 137-38. This suggestion is apparently
based upon the notion that the middle combined with the verbal idea sug-
gest the notion of asking for oneself with selfish interests.
Faith, (Chicago: Moody Press, 1979), p. 248.
3 Leddusire, "Middle Voice," p. 129.
4 Zerwick, BG, p. 76.
5 However, using this passage as a parallel to James 4:2 is only
supportive and does not establish the distinction as always valid. For
a different viewpoint, see William Hendriksen, Exposition of the Gospel
according to Mark (Grand Rapids: Baker Book House, 1975), p. 240.
6 Zerwick, BG, p. 76.
element.1 Leddusire offers a paraphrase which bears out the voice dis-
tinctions. "You do not have because you are unaffected by asking. When
you do ask, you are without results because your interest in asking is
undesirable, namely to squander with your sensualities.2 Using genera-
tive transformational grammar, he concludes that the persistence of overt
markers in a system where the contrasts are demonstrably productive point
to distinction. However, the interpretation of this assertion in terms
of traditional grammar is uncertain. For he must ultimately depend upon
context to give two different meanings to the middle of ai]te<w, i.e.,
because you are unaffected by asking (dia> to> mh> ai]tei?sqai u[ma?j) and
because your interest in asking is undesireable (dio<ti kakw?j ai]tei?sqe).
Furthermore, the validity of the suggestion because you are unaffected
by asking is very dubious. Is James stating that if his readers are
affected by their asking, then their requests will be answered? How is
one to be affected by his own asking? If this was the crucial point of
the condition, it would seem that James would make plain the answers to
such questions. Thus, this suggestion appears to be forced and unnatural.
This alternation of voices in James 4:2,3 has also been viewed
as simply an arbitrary interchange.3 Yet, as Moulton suggests, it is
difficult to understand how a writer like James could permit so
purposeless a freak as this would be.4 Perhaps on the basis of style
1 Hiebert, James, p. 248.
2 Leddusire, "Middle Voice," p. 131.
3 GOEL, p. 166; Simcox, The Language of the New Testament, p. 95;
Nunn, A Short Syntax of New Testament Greek (
University Press, 1912), p. 64.
4 GNTG 1:160. Although he argues against an arbitrary inter-
change, he concludes this usage is an extinct subtlety.
the middle forms were adopted to balance the two active forms ai]tei?te
and ou] lamba<nete.1 Yet this also is a tenuous suggestion, for such
stylistic usage of voice does not appear elsewhere in James.2
In view of this controversy and lack of strong support for either
position, Adamson correctly observes that no certain distinction has
been established between the active and middle in this passage.3 Yet
there are also no cogent reasons which eliminate the possibility of the
middle conveying an intensification between the subject and its verbal
action.4 This context suggests the possibility that the intensification
may be the personal interest of the subject in the request. Thus, this
passage is certainly not irrefutable evidence that the active and middle
voices of certain verbs are used interchangeably, nor vice versa.5
1 John 5:14, 15
Parallel in difficulty are the five occurrences of ai]te<w in
1 John.6 Within two verses there is a variation of middle, middle, and
1 James B. Adamson, The Epistle of James, NICNT, ed. F. F. Bruce
(Grand Rapids: Wm. B. Eerdmans Publishing Co., 1976), p. 169.
2 An examination of the flow of verbs according to voice in James
has not revealed another sequence of middle-active-middle-active or any
3 Adamson, James, p. 169.
4 For opposing view, see TDNT, s.v. “ai]te<w,” by Gustov Stahlin,
1:192. Although he states that there is no option but to explain this
voice variation in James in terms of the formal structure of the sen-
tence, his arguments are really only applicable to Mayor's suggestion.
5 Turner, Grammatical Insights, p. 164. Hence his assertive
conclusions for interchangeability need to be more balanced. Even
BAGD, p. 25, concludes that the middle and active only seem to be used
6 1 Jn 3:22, 5:14, 15, 16. Also the twelve occurrences of ai]te<w
in the Gospel of John display voice variation and present difficulties
active. "And this is the confidence which we have before Him, that, if
we ask (ai]tw<meqa) anything according to his will He hears us. And if we
know that He hears us in whatever we ask (ai]tw<meqa), we know that we have
the requests which we have asked (^]th<kamen) from Him" (1 John 5:14, 15).
Certainly in this passage the qualifying phrase kata> to> qe<lhma does not
seem to permit any self-interest to be involved. On the basis of the
usage of ai]te<w in contexts of business dealings where the middle may add
the nuance that one has the right to ask, it is suggested that this
difference in meaning is apparent and certainly seems to be intended.
Why should the two middle forms that are used here not include
this right? Does the phrase 'according to his will" (qe<lhma, what
God has willed and has made known as being willed by him) not imply
a certain right for our asking?1
However, the context does not support this nuance. The requi-
sitioning in prayer is the same in both ai]tw<meqa and ^]th<kamen without
adverbal modifiers as in James 4:2, 3. Although perceiving no difference
in meaning, two suggestions attempt to account for the variation in form
in this passage. First, the cognate accusative ai]tei?n ai]th<mata in the
active voice is understood as a periphrasis for the middle ai]tei?sqai.2
Second, it is suggested that in Johannine usage the active is used with
the accusative.3 These notions, however, appear to be inadequate,
1 R. C. U. Lenski, The Interpretation of the Epistles of St.
1966), p. 533.
2 David Smith, "The Epistles of John," in vol. 5 of Expositor's
ed. W. Robertson Nicoll, (
Publishing Co., 1979), p. 197. He does note a difference of meaning in
3 Robert Law, The Tests of Life: A Study of the First Epistle of
Rapids: Baker Book House, 1968), p. 406. An exception is John 11:22.
Also the usage of the active of ai]te<w in James does not follow this pat-
tern, although the difference may be accounted for simply on the basis
of different authorship.
for they confuse the notion of transitiveness with that of voice.
Therefore, in this passage neither a difference of meaning
between active and middle is discernible, nor does the difference
appear to be satisfactorily explained in teams of transitiveness.
While there may be a semantic distinction of voice regarding ai]te<w
in James, none is discernible in 1 John. Thus, one should be wary
of broad generalizations regarding voice distinctions, even with a
specific verb, apart from an examination of each individual context.1
Parallel Synoptic Passages
Striking evidence for the notion of interchangeability of
middle and active without semantic difference may be derived from
parallel synoptic accounts. Whereas one author uses the middle voice,
another author employs the active voice in the same verb while describ-
ing the same event.
Matthew 26:23; Mark 14:20
The particular detail with voice variation is in the significant
description by Jesus of the traitor.2 Mark uses the middle voice and
Matthew uses the active. "He who dips (o[ e]mbapo<menoj) with me into the
dish" (Mark 14:20). "He who dips (o[ e]mba<yaj) his hand in the dish with
me" (Matt, 26:23). Yet not only does the voice vary, but also the tenses
1 For example, see DNTT, s.v. "Prayer," by H. Schonweiss,
2 This specific detail is omitted in the Lukan and Johannine
accounts. Furthermore, e]mba<ptw does not occur elsewhere in the NT or
LXX, apart from the textual variant at John 13:26.
are present and aorist, respectively. Any intended difference of
meaning by either writer in his use of tense is not readily discernible.1
However, a lexical citation of these passages gives dip for the active
and dip for oneself as the middle.2 This additional nuance in the middle
is in accord with Gould's suggestion that Mark does not mean to indicate
the traitor, but only to emphasize the treachery of the act.3 But this
emphasis may be understood apart from any contribution of voice.
Matthew 19:20; Mark 10:20; Luke 18:18
The rich young ruler's response to Jesus concerning the command-
ments involves the use of fula<ssw. Whereas Matthew and Luke both use
the aorist active e]fu<laca, Mark uses the aorist middle. "Teacher, I
have kept (e]fula<camhn) all these things from my youth" (Mark 10:20).
Leddusire, finding a semantic difference, attempts to explain this in
terms of a dative middle model which has the inference of an affected
subject. He attempts to gather further contextual support from the
young ruler's questioning of Christ.
The exegetical distinction is also supported in the context, which
follows the original question in Mark 10:17, "What (else) must I
do?" On the other hand, the active sentence of Matthew is in
1 R. C. H. Lenski, The Interpretation of St. Mark's Gospel
(Minneapolis: Augsburg Publishing House, 1961), p. 616. He views both
the aorist and present as timeless tenses with any intended difference
of meaning as unlikely.
2 BAGD, p. 254. However, no difference is stated in LSJ, p. 539.
The shift of tense from the present of o[ e]mbapto<menoj to the aorist
o[ e]mba<yaj cannot be accounted for by deponency. The verb e]mba<ptw is not
a middle deponent form for the present but an active form for the aorist.
3 Ezra P. Gould, A Critical and Exegetical Commentary on the
according to St. Mark,
ICC ed. C. A. Briggs, et al. (
Charles Scribner's Sons, 1896), p. 262. His suggestion is not based on
simply the voice difference.
answer to the question in Matt 19:16, "what is a good action I can
perform?" and can be paraphrased as "why, I've already done that."1
Yet the following three questions posed by the young man and
directed to Christ have little, if any, difference.
1. "Teacher, what good thing shall I do that I may obtain eternal life"
2. "Good teacher, what shall I do to inherit eternal life?" (Mark 10:17)
3. "Good teacher, what shall I do to obtain eternal life?" (Luke 18:18)
In fact, the only difference between the question in Mark and
Luke is the use of e@xw rather than klhronome<w in the i!na clause. Thus,
the cause of Mark's use of the middle e]fula<camhn is not to be found in
this question. Nor is the suggestion that Matthew and Luke independently
corrected Mark's use of e]fula<camhn particularly cogent.2 Brown cau-
tiously concludes that, while the middle may have the same force as the
active, it may also mean guard for oneself, store, or be careful.3 If
Mark intends to clarify that the young ruler has emphasized his guarding
of the commandments in relation to himself, it is extremely difficult to
detect this from contextual clues. The contexts, including specific
details, are nearly identical by each author.
Matthew 26:51; Mark 14:47
These two writers, while reporting a specific detail of a single
event, selected different voices for its transmission. When describing
1 Leddusire, "Middle Voice," pp. 136-38.
2 Vincent Taylor, The Gospel according to St. Mark, 2d ed.
(London: Macmillan and Co., 1966), p. 428. He not only assumes the
primacy of a Marcion source, but also assumes misuse of the middle by
Mark. However, if the middle is interchangeable with the active, it is
simply disused, not used incorrectly (Simcox, Language of the New
Testament, p. 96).
3 DNTT, s.v. "Guard," by C. Brown, 2:134.
the drawing of the short sword from its scabbard, Matthew uses
a]pe<spasen, but Mark employs spasa<menoj. Again, it is Mark who consis-
tently uses the middle form when there is a voice form difference.1
Matthew's use of the prefix a]po> with the verb does not add any additional
significance to spa<w.2
From these synoptic passages, several factors emerge. First,
various verbal features may vary without any semantic significance.3
These include the choice of a particular verb, the selection of a speci-
fic tense of the same verb, the selection of a specific voice of the
same verb, and the addition of a prefix to the verb. Mark has been the
only author known who consistently uses the middle when parallel synoptic
accounts have the active. Thus, it appears that Mark may simply have a
stylistic preference for the middle without an intended difference of
meaning, compared to the active, for no intended difference is
Additional support for the theory of voice interchangeability
has been gathered from sentences which, although contextually disparate,
1 Although Luke does not include this detail, John includes it
with the usage of a different but synonymous verb e!lkw.
2 BAGD, pp. 98, 761.
3 For other conspicuous grammatical differences without apparent
semantic significance, see Alfred Plummer, An Exegetical Commentary on
the Gospel according to St. Matthew, 3d ed. (London: Robert Scott, 1911),
4 For a different conclusion, see Leddusire, "Middle Voice," p.
135. While recognizing Matthew's stylistic preference for the active
as undisputed, he views this fact as irrelevant to the theory of voice.
use the same verb. A verb in fts active voice form is paired with an
occurrence of its middle form in a different context.1
The perfect active infinitive eu[rh<kenai is cited as having no
semantic difference as compared to the aorist middle participle
eu[ra<menoj (Rom 4:1, Heb 9:12).2 However, this assertion is wholly
arbitrary and subjective. Appropriate criteria for the establishment of
voice interchange, i.e., parallelisms, contextual similarity or identity,
and stylistic preferences, are lacking.3 The middle voice of eu[ra<menoj
can be clearly distinguished from the active.4
In a similar vein, the active of u[stere<w in Hebrews 4:1, 12:15
is viewed as possessing exactly the same significance as the middle in
Romans 3:32.5 Again, the same objections regarding eu[ri<skw are appli-
cable to this methodology. No evidence is cited by either Winer or
Simcox to support their assertions.
1 These same. verbs are also acknowledged by others to have
semantic difference according to voice form. For example, note the
lexical listings of fai<nw in BAGD, pp. 851-2.
2 Simcox, Language of the New Testament, p. 96.
3 This problem is further compounded by the fact that non-finite
forms, especially infinitives, do not always reflect the force of the
voice. Gildersleeve, Syntax of Classical Greek 1:63; Leddusire, "Middle
Voice," p. 42.
4 Brooke F. Wescott, The Epistle to the Hebrews (
millan Co., n.d.; reprint
5 Simcox, Language of the New Testament, p. 95; Winer, A Grammar
of the Idiom of the New Testament, p. 260. Winer more generally con-
cludes that the middle and the active of this verb are always synonymous
Using Additional Verbs
Although they are cited without specific passage indicators,
the following verbs have been purported as having interchangeable voice
forms without semantic distinction: la<mpw; o[ra<w; se<bw; and fai<nw.1
From these passages and specific verbs, it appears that the
assumption of voice interchangeability has a very weak foundation. Sup-
portive passages are tenuous and infrequent. It is possible, although
not probable, that in each of the cited passages the middle voice conveys
in some degree or manner an intensification of the relationship between
the subject and the verbal action. However, the voice interchange in
parallel synoptic passages renders this as improbable. Yet each specific
passage must be examined in light of its own contextual factors, and
broad generalizations promoting interchangeability should be avoided.
For while the middle and active of ai]te<w appear to be semantically dis-
tinct in James 4:2, 3, this is not the case in 1 John 5:14, 15.
Active for Middle
The assumption that the active is used for the middle as sup-
portive of interchangeability rests on several slightly different
Based on Similarity of Meaning
Concerning classical usage, Smyth observes that the active is
often used for the middle when it is not of practical importance to
1 Thompson, A Syntax of Attic Greek, p. 160; James T. Allen, The
First Year of Greek (New York: Macmillan Co., 1932), p. 310; Winer,
A Grammar of the Idiom of the New Testament, pp. 256-57. As in the
previous examples, no solid evidence is cited.
mark the interest of the subject in the action. The active implies what
the middle expresses.1 Regarding Attic usage in particular, it is noted
that the active is used like the middle.2 Inarguably, a significant
difference of meaning between the active and middle forms of the verbs
cited in their examples is not evident.3 However, similarity in meaning
does not necessarily establish identity of usage in general. As Turner
observes, the verbal idea inherent in certain verbs is not significantly
expressed as a difference in either active or middle.
For practical purposes, it mattered very little whether the
active or middle voice was used with verbs of a certain type. "I
make a request" is active, but is not profoundly different from the
middle, "I make a request for myself." It defines the idea more
narrowly (emphasis mine), but in normal conversation, either active
or middle would do.4
But even as Turner recognizes, this does not mean that no subtle
nuance may be intended. Thus, rather than assuming that the active is
used for the middle, it seems better to view this phenomenon as a
result of the verbal idea. Certain verbal ideas do not have a signifi-
cant semantic shift in active to middle, but subtle nuances may be
Based on Classical Precedent
Some verbs are thought to appear in the active where the middle
would be expected in classical Greek.5 The most notable example is
poie<w with a verbal noun. In classical Greek, there are numerous
1 Smyth, Greek Grammar, p. 393.
2 Thompson, Syntax of Attic Greek, p. 167.
3 metape<mpw, dhlo<w, dida<skw, metaxeiri<zw, bia<zw, pare<xw,
4 Turner, Grammatical Insights, p. 163.
5 GNTG, 3:56.
differences between poiei?n and poiei?sqai with verbal nouns in which the
active gives the literal side "to fashion," "to bring about," whereas
the middle serves to form a periphrasis with the verbal noun for the
corresponding verb.1 This periphrasis, composed of poiei?n in the middle
voice plus a noun denoting action as an object, is equivalent to a
simple verb.2 However, lo<gon poiei?sqai (to make a speech) may correspond
to legei?n, but it is not the same as lo<gon poiei?n (to compose a speech).
Similarly, o[do>n poiei?sqai (to make one's way) may correspond to o[deuei?n,
but this is not the same as o[do>n poiei?n (to construct a road). Thus,
using this criterion, the middle would be expected in Mark 2:23, but in
fact the active occurs. "And his disciples began to make their way
(o[do>n poiei?n) while plucking the heads of grain" (Mark 2:23). Yet, this
assumption that the classical distinction is lost may be challcnged.3
A possible explanation is that the disciples began to make a way, i.e.,
to open a path, by plucking the ears of corn.4 But this cannot be
maintained as an inviolable rule, for the LXX clearly uses o[do>n poiei?n
1 Gildersleeve, Syntax of Classical Greek, p. 69; Smyth, Greek
Grammar, p. 391.
2 BG, pp. 72-73. Examples cited as evidence include porei<an
poiei?sqai for poreu<esqai, mnei?an poiei?sqai for memnh?sqai. Also see
James L. Boyer, "Notes on 2 Peter and Jude" (Winona Lake, IN, 1977),
p. 10. Perhaps the middle sense of bebai<an poiei?sqai< should not be
pressed, since Greek idiom in classical Greek required the middle. In
the NT both active and middle forms of poiei?n are used in this peri-
3 Heinrich A. W. Meyer, Critical and Exegetical Handbook to the
Gospels of Mark and Luke, vol. 2 trans. Robert E. Wallis in Meyer's
Commentary on the New Testament, rev. and ed. William P. Dickson
1979), p. 33; Alexander B. Bruce, "The Synoptic Gospels," in vol. 1 of
ed. W. Robertson Nicoll (
B. Eerdmans Publishing Co_, 1979), pp. 354-55.
4 Meyer, Critical and Exegetical Hand-book to the Gospels of
Mark and Luke, p. 33.
in the sense of to make one's way, to journey. "Then the man departed
from the city, from Bethlehem of Judah, to dwell wherever he might find
a place, and he came to the hill district of Ephraim to the house of
Micah as he made his journey (tou? poih?sai th>n o[do>n au]tou?)" (Judg 17:8).1
Thus, the criterion of a classical precedent may be used to establish
either view, and it is a tenuous standard for the determination of voice
interchange without semantic distinction. Even if o[do>n poiei?n means to
make one's way in Mark 2:23, this only demonstrates a difference of
classical and koine usage. It does not establish the notion of inter-
changeability in the NT.
Based on Different Construction
In the NT, a verb in the active voice with a reflexive pronoun
is numerically predominant over the direct reflexive usage of the middle
voice.2 These two different constructions have been equated in terms of
semantic significance in the NT.3 In Luke 16:9, e[autoi?j poih<sate might
have been fully expressed by one word, poih<sasqe.4 Similarly, the dif-
ference between prose<xete e[autoi?j and fula<ssesqe is viewed as minimal
in Luke 12:1, 15.5 Yet, Robertson's conclusion that the use of the
1 Alfred Rahlfs, ed. Septuginta, vol. 1 (Stuttgart: Wurttem-
bergische Bibelanstalt, 1935), p. 476.
2 Alfred S. Geden and William E. Moulton, eds. A Concordance to
the Greek Testament, 5th ed. rev. Harold K. Moulton (Edimburgh: T. and
T. Clark, 1978), pp. 240-44.
3 BGHG, 2:398.
4 Samuel Green, Handbook to the Grammar of the Greek New Testa-
numerical dominance of active voice with reflexive pronoun in koine
seems to indicate a loss of the directly reflexive sense in most cases.
5 GNTG 1:157. Perhaps the reflexive construction is slightly
reflexive pronoun with the active bears more sharply the reflexive
relation than the mere middle has more justification.1 For as early as
Homer, the reflexive forms are occasionally used with the middle to more
clearly bring out the reflexive notion.2 Regardless of how closely the
two constructions are identified in meaning this does not establish the
notion of interchangeability. For the active voice per se is not equated
with the middle, but rather the active with reflexive pronoun.
Therefore, in summary, the assumption of active for middle
usually stands without warrant. Certain verbal ideas may be signifi-
cantly different in their active as compared to middle voices, but this
is due to the nature of the verbal idea. Also the appeal to classical
usage is a two-pronged argument that may validate either position.
However, even if the active is used where a middle might appear more
appropriate in classical usage, the only fact established is that of a
difference between koine and classical. The notion of interchangeability
in the NT has not been supported. Finally, a difference of construction
with identical or very similar meaning also fails to support voice
interchange. Jannaris' conclusion that the use of the active instead of
the middle occurs times without number is unwarranted.3
Passive as Middle
The aorist passive of some active verbs may have a reflexive or
middle sense.4 Whereas fai<nw means show, e]fa<nhn showed myself,
1 GLHR, p. 802. Also, see BG, p. 75.
2 Gildersleeve, Greek Syntax, p. 68. In the NT, note Acts 7:21,
20:24, 1 Tim 3:13; Titus 2:7. However, this phenomenon usually occurs
3 HGG, p. 364. His numerous NT illustrations usually involve poie<w.
4 Smyth, Greek Grammar, p. 222. He identifies these verbs as
appeared. The same type of semantic shift is true of eu]frai<nw, kine<w,
and xai<rw. However, this phenomenon appears to be adequately accounted
for by the historical development of the qhn aorist. The passive idea
was not always the original sense, and hence, in NT times, the passive
idea is not perceptible in these verbs.1 This does not support voice
interchange in the sense that the middle and passive voices are used
interchangeably. Instead, these passives are simply used with a mild
reflexive sense. The middle and passive of the same verb do not occur
in parallel passages with semantic identity.
In this section, except for deponents, fall the usages of the
middle voice which do not overlap in meaning with the active and passive
voices. In order to analyze the various usages, it is a matter of
convenience to refer to the divisions of the middle voice constructed
by grammarians. However, these divisions appear, as Robertson main-
tains, more or less arbitrary and unsatisfactory.2 Almost every
grammarian differs to a certain extent in his terminology and categori-
zation, for the Greeks themselves did not need or possess such divisions.
Grammarians have listed as few as two to as many as nine categories.3
Furthermore, Green calls the reflexive usage direct or indirect, whereas
1 GNTG 1:161.
2 GLHR, p. 806. Also, see MGNT, p. 158.
3 HGG, pp. 360-61. He places all usages in either a directly
reflexive or an indirectly reflexive category. For nine categories,
William W. Goodwin, Greek Grammar, rev. Charles B. Gulick
Ginn and Co., 1930).
Brooks and Winbery classify the same phenomenon as dynamic or
intensive.1 However, this is not an indictment against grammarians,
for the categories are erected for analytic and didactic purposes. Even
Dana and Mantey, who employ a taxonomic approach, offer the following
An analysis of the uses of the middle is of necessity more or
less arbitrary. No rigid lines of distinction can in reality be
drawn. Distinctions there are, however, and the following analysis
is proposed as indicating the main lines of difference.2
Furthermore, when recognizing distinct nuances of usage of the
middle voice, it is helpful to employ a distinctive term to describe the
particular phenomenon of language. However, by, no means does this mean
that these categories are an essential feature of the fundamental signi-
ficance of the middle voice. The middle voice per se only relates an
intensification of the relationship between the subject and the action
expressed by the verb. The degree or manner of intensification may be
mild or acute, and the determination of the intensification is in terms
of a particular context and the meaning of a verb.3 Thus, these cate-
gories are of usage and not of features inherent in the middle voice
Since the categories are defined differently by grammarians, a
somewhat arbitrary selection of the terminology and categorization of
one author will be consistently employed in order to avoid confusion.
As Robertson's six categories are generally defined and thoroughly
1 Green, Handbook to the Grammar of the Greek Testament, p. 292.
This taxonomical confusion repeatedly occurs among grammarians.
2 NGNT, p. 158.
3 Ibid., p. 158.
illustrated, they will conveniently serve as the basis for an analysis
In the directly reflexive usage, the intensification of the
subject to verbal action is such that the action is directly upon or to
the subject. Although Jay denies this category, and Moulton only accepts
one possible example in a]ph<cato, Robertson offers over twenty illustra-
tions.2 However, over one-half of these examples may also be identified
as passive and are questionable.3 Thus, although a directly reflexive
sense does occur in the NT, the number of occurrences is extremely
Causative or Permissive Middle
The labeling of the middle voice as causative appears to be
unwarranted.5 The active voice is also designated as causative, but as
both Robertson and Jannaris observe, this feature is not due to the
voice.6 In addition, this feature is common to all languages.7 If
transitiveness is to be properly separated from the notion of voice,
1 GLHR, p. 106. Even Robertson follows these divisions merely
2 Ibid., pp. 806-08.
3 For example, note the verbs pota<ssesqe, dogmati<zesqe, and
4 MGNT, p. 158.
5 Although Robertson does not explicitly define the term
causative, his citation of Gildersleeve gives the impression that he is
following Gildersleeve's definition.
6 GLHR, p. 801.
7 HGG, p. 359.
then so also is causation.1
Neither is the permissive label particularly lucid. The per-
missive sense of the middle is considered as closely allied to the
causative and approaches the passive.2 This permissive middle has been
more clearly defined as representing the agent as voluntarily yielding
himself to the results of the action, or seeking to secure the results
of the action, or seeking to secure the results of the action in his own
interest.3 Simply stated, the action takes place by order or with per-
mission of the subject.4 Thus, the intensification of the relationship
between subject and verbal action is such that the subject permits or
allows the action. Again, it should be noted that this is derived from
the context and the root idea of the verb. Dani<sasqai and misqw<sasqai
appear to be valid examples of this usage (Matt 5:42; Matt 20:1).5
In this usage the subject is represented as doing something for
or by himself. This indirect usage is quite varied and abundant in the
NT. Often the subject is merely highlighted as the doer of the action.
This, along with the dynamic category, is very vague, and perhaps the
two should be combined. For even Robertson finally concludes concerning
this category that each word and its context must determine the result.6
1 Thompson, Syntax of Attic Greek, p. 162. Also see GLHR, p. 809.
The causative idea in a]nakefalaiw<sasqai ta> panta> e]n t&? Xrist&? is not
due to the voice, but to the verb itself (Eph 1:10).
2 GLHR, p. 809.
3 MGNT, p. 160.
4 Winer, A Grammar of the Idiom of the New Testament, p. 254.
5 GLHR, p. 809.
6 Ibid., p. 809.
In fact, the exact relation of the indirectly reflexive usage must be
perpetually varied if the sense of the middle is to be appropriate to
the particular example.1
An interchange of effort between the members of a plural subject
may be expressed by the middle voice.2 This usage appears to be semanti-
cally equivalent to the active voice with a reciprocal pronoun.3 The LXX
quotation of diemeri<santo e[autoi?j from Psalm 21:19 is given as
diemeri<santo e[autoij in John 19:24, but is only stated as diemeri<santo
without the reciprocal pronoun in Matthew 27:35. Therefore, in Matthew the
middle appears to be clearly used in either a reciprocal or distributive sense.
In this usage both the pronoun and the middle occur.4 This
redundance also exists in classical Greek, and it may represent more
clearly the reflexive force in some cases.5 Overlap within these cate-
gories is apparent, for diemeri<santo e[autoi?j, although being reciprocal,
also falls within this class (John 19:24).
Dynamic or Deponent
Whereas certain grammarians have a separate category for dynamic
and for deponent, Robertson combines them.6 Gildersleeve's remark that
1 GNTG, 1:157.
2 MGNT, p. 160.
3 GLHR, p. 810.
4 Ibid., p. 811.
5 Gildersleeve, Syntax of Classical Greek, 1:68.
6 GLHR, p. 811. Also see Thompson, Syntax of Attic Greek, p.
this is the drip-pan or pande<kthj middle that is put at the bottom to
catch the drippings of the other uses clearly demonstrates the diffi-
culty of applying a label to every usage of the middle.1
However, it is important to recognize the phenomenon usually
described by the term deponent. Deponent verbs have been defined as
verbs which have no active forms, but only middle or passive forms with
active meaning.2 However, this definition is inadequate for advanced
students because certain verbs, especially in the future tense, have
both an active and a middle form with the middle voice form performing
an active voice function. Both a]kou<sw and a]kou<somai are found in the
NT, with a]kou<somai having an active voice function (Matt 12:19; Acts
3:22).3 Rather than a facet of voice interchange, this phenomenon is
closely parallel to verbs which are deponent only in the future. Thus
the distinctive feature of a deponent is that its voice form is
different from its voice function.4 The active voice form may also
occur when a middle form is deponent, although this is usually not the
The identification of a deponent middle form is not simply
limited to a lexicon. Whereas Thayer, Abbott-Smith and LSJ have an
1 Gildersleeve, "Stahl's Syntax of the Greek Verb," p. 277.
Macmillan Co., 1951), p. 61.
3 GNTG 1:154
4 MGNT, p. 163.
5 For additional verbs exhibiting this feature, see GNTG 1:154-55.
active form for proxeiri<zw, LPGL and Sophocles have a deponent lexical
listing.1 BAGD lists it as active but observes that it is only middle
deponent in the included literature.2 The extent of the literature sur-
veyed is a contributing factor in identifying a deponent middle. However,
usage in the particular contextual environment is the key indication.3
Two areas of usage of the middle voice have been investigated.
First, regarding the phenomenon of voice interchange without semantic
difference, there is scant supportive evidence in the NT. An investi-
gation of parallel synoptic passages as well as key individual texts
does reveal voice interchange without semantic distinction as occurring.
However, rather than being a general rule, this phenomenon must be
determined per individual context.
Regarding the divisions of usage, they are not derived from any
inherent feature of the middle voice per se. Contextual factors com-
bined with the verbal idea are the foundation upon which these divisions
have been erected. Naturally, therefore, they vary from grammarian to
grammarian and are somewhat arbitrary. Yet, it is important to recognize
the category of deponent, i.e., one whose distinctive feature is an
1 Joseph H. Thayer, Greek-English Lexicon of the New Testament
(Grand Rapids: Zondervon Publishing House, 1975), p. 554; Georg Abbott-
A Manual Greek Lexicon of the New Testament (
Scribner's Sons, 1922), p. 391; LSJ, p. 1541; Evangelinus A. Sophocles,
Lexicon of the Roman and Byzantine Periods, vol. 2 (
2 BAGD, p. 724. Also see lumai<nw, parrhsia<zw, e]pilamba<nw,
e]pektei<nw, strateu<w in the various lexicons.
3 Concerning the problematic identification of pau<somntai as a pos-
sible deponent in 1 Cor 13:8, see Charles Smith, Tongues in Biblical
Perspective (Winona Lake: BMH Books, 1972), pp. 83-84.
active meaning with a middle form. Since for certain verbs the issue of
deponency is not clear, further lexicography needs to be performed.
TRANSLATION AND INTERPRETATION
Before suggesting general guidelines, it is appropriate to
submit warnings that should remove artificial prescriptive rules.
Not a single grammarian has been encountered who advocates the
translation of every middle. Instead, they have appropriately warned
against overtranslating the middle voice by attempting to express every
single shade of meaning by an English word or phrase.1 The variation of
the middle form may be too minute for translational discrimination.2
Stahl's attempts to translate the middle are cogently corrected by
We translate i]dei?n to see and i]de<sqai to see with one's own
eyes; an overtranslation as o]fqalmoi<sin o[ra?n shows, but if there is
such virtue in i]de<sqai, why not in i]doma<noj? Ah! the verse. Like
the rest of us, Stahl has to go into bankruptcy. Translation will
Similarly, Smyth submits that the force of the middle in
a]kou<esqai, ti<masqai, a]riqmei?sqai, and a]porei?sqai cannot be reproduced
in translation. In some cases, it may not have even been felt.
2 GLHR, p. 804.
3 Gildersleeve, "Stahl's Syntax of the Greek Verb," p. 278.
Against the definitive, exhaustive approach of erecting rigid
rules in any language stands the timely warning of Meyer-Myklestad.
Within the limits imposed by the syntactic possibilities of a
language, the speaker is a free agent: grammar cannot compel him to
think this way or that. The sentence is instructive in that it shows
the impossibility of prescriptive rules in grammar.1
Hence, it reasonably follows that no fixed rigid rule can be
maintained for the translation of a particular use of the middle voice.2
If the categories of usage themselves overlap and are somewhat arbitrary
and indistinct, how can a fixed rule be erected for that category?
Instead, each particular occurrence must be analyzed separately.
In view of the difficulty involved in interpreting and trans-
lating many occurrences of the middle voice, it appears sound to conclude
with Moule that as a rule it is far from easy to come down from the
fence with much decisiveness on either side in an exegetical problem if
it depends on voice.3 The assertion that the middle voice of pau<sontai
demonstrates that tongues are no longer extant today is highly
gratuitous (1 Cor 13:8).4 It is possible to reach a valid conclusion
based on partially erroneous exegetical reasoning since that conclusion
1 Johannes Meyer-Myklestad, An Advanced English Grammar for
Students and Teachers (New York: St. Martin's Press, 1969), p.
2 GLHR, p. 810.
3 Moule, An Idiom Book of New Testament Greek, p. 24. However,
for four arguments against this conclusion, see Leddusire, "Middle
Voice," p. 135. Since his presupposition that generative transforma-
tional grammar is the only adequate framework for voice study problems
has been questioned, his conclusions are unconvincing.
4 Again, for cogent argumentation, see Smith, Tongues in Biblical
Perspective, pp. 83-84.
may be demonstrably valid via other argumentation. But this does not
condone improper methodology and unwarranted dogmatism that will
normally yield unsupportable results.
Authorial and Geographical Variations
Moulton's conclusion that usage inevitably varied in different
localities and between different authors appears sound.1 From the
parallel synoptic passages, it has been suggested that Mark's use of the
middle compared to the active in other passages may simply be a
stylistic feature. Furthermore, the voice interchange in James 4:2, 3
may be explained as the writer's stylistic variation adopted to balance
the two active forms.2 Perhaps also the usage of the middle would vary
with the writer's Greek culture.3
Insistence on Classical Greek Distinctions
It appears hazardous to agree with the conclusion that the
system of voices in general remained the same in the Hellenstic period,
including the NT, as in the classical period of the language.4 To the
other extreme, Turner concludes that NT writers are not happy in their
understanding of the middle voice according to classical standards.5
One of the principal characteristics of NT Greek in general is the
1 GNTG 1:159.
2 Adamson, James, p. 169. However, that is not the position
adopted in this paper.
3 GNTG 1:159.
4 GOECL, p. 161.
5 Turner, Grammatical Insights, pp. 106-7.
absence of classical Greek standards.1 Although a middle form of a verb
may have had a distinctive sense in classical Greek, this meaning should
not be automatically carried over into the NT.2
Only two basic guidelines emerge from this study that appear to
Each particular occurrence of the middle voice must be weighed
in terms of the historical development of the verb, primacy of context
and the idea itself. These factors determine not only if there is any
intensification between the subject and the action expressed by the
verb, but also the degree and manner of intensification. Although one
may not always be able to clearly express the middle voice by an English
translation, one can seek to acclimate oneself to its mental atmosphere
and feel its force by repeated exposure in different contexts with
different verbs.3 Moulton's suggestion that "He pardoneth" could be
used to represent a]fi<etai, whereas "He pardoneth" expresses a]fi<hsi,
would be valid only if the particular context indicated that this was
the emphasis. The same is true for Dana and Mantey's suggestion for the
use of italics.4
1 DNTT, s.v. "Presuppositions and Theology in the Greek New Tes-
tament," by Murray J. Harris, 3:1171-1215. Many of his observations do
not simply regard prepositions but the language as a whole.
2 Observe poiei?n o[do>n in the discussion of Mark 2:23 by Bruce,
"The Synoptic Gospels," pp. 354-55.
3 MGNT, p. 157. In addition, general guidelines and an elementary
procedure for translating Greek into English are offered by Gideon and
4 MGNT, p. 159. The Greeks employed the middle where we must
resort to italics.
As it is difficult, if not impossible, to translate without
interpretation, the preceding suggestions are applicable here. In addi-
tion, Blass' conclusion that on the whole the NT writers were perfectly
capable of preserving the distinction between the active and the middle
appears to be sound.1 Thus, although there is some usage which may be
synonymous in meaning among the voices, voice interchange is an infre-
quent phenomenon. The probable exegetical significance of a true middle
as dictated per context should not be overlooked.
1 Blass, Grammar of New Testament Greek, p. 186.
The grammatical category of voice is the relationship between
the subject of a sentence and the action expressed by the verb. For the
sake of clarity and consistency, it is advantageous to define the three
Greek voices in terms of this relationship. The notion of general
reflexivity, although an apparent feature of the middle voice, does not
elucidate the nature of this relationship. General reflexivity is vague
and imprecise, and does not considerably aid one's comprehension. In
addition, the concepts of middle signification and transitiveness are
either inadequate or irrelevant regarding voice meaning. Although the
concepts of special advantage and subject participation in the results
may be involved at times, these ideas are not inherent to voice itself.
Historical argumentation and usage remove the idea that the middle voice
is middle in meaning between active and passive. Instead, a basic notion
of the middle voice as an intensification in some manner or degree of
the relationship between the subject and the action expressed by the
verb serves as a valid guideline. The precise nature of this intensifi-
cation between subject and verbal action is not indicated by the middle
voice per se. The nature of the intensification must be derived from
the context, the historical evolution of the verb, and the verbal idea
itself. Thus, even though this basic concept regarding the middle voice
occurs in the majority of NT true middles, it may be absent or modified
as indicated by these factors.
Concerning the controversy regarding voice interchange without
semantic distinction, the phenomenon does appear to exist but in a very
limited number of cases. An investigation of parallel synoptic passages
and key texts with voice interchange reveals that no apparent distinction
is intended in certain cases. However, no general rule of thumb is
available regarding this voice variation. For in one passage an inten-
ded semantic shift can be detected, but in another passage no semantic
distinction is apparent.
Regarding the divisions of the middle voice, they are not
derived from the middle voice per se. Contextual factors and the verbal
idea are the foundation upon which these categories have been erected.
The divisions are not rigid and definitive, but are somewhat arbitrary
and overlap. The division of deponency is the most important category
which includes middle voice forms with an active function. The identi-
fication of a deponent is not simply via lexicons, but in certain
questionable cases further lexicography is needed.
Several warnings regarding translation and interpretation have
emerged from this study. The middle voices cannot always be expressed
by means of translation. Certain verbal ideas per se do not suggest
that this is possible, and apparently the Greeks did not always intend a
major difference. At times the variation of the middle from the active
is so minute it is difficult to know if one has properly recognized an
intended distinction. In view of this, it is difficult to be decisive
in an exegetical problem if it depends on voice.
Also an author may use a specific voice as a stylistic feature,
but this is not a general rule. However, it does warn against
establishing principles without considering possible authorial tendency
Finally, classical Greek distinctions per se should not be used
to determine NT usage. Examples contrary to classical usage do appear.
A distinctive classical meaning for a middle voice should not be
automatically carried over into the NT.
Edwin. Johannine Grammar.
Abbott-Smith, George. A Manual Greek Lexicon of the New Testament.
Adamson, James B. The Epistle of James. NICNT. Edited by F. F. Bruce.
Alford, Henry. Alford's Greek Testament, an Exegetical and Critical
Commentary. vol. 4.
James T. The First Year of Greek.
According to St. Matthew. ICC. 3d ed. Edited by C. A. Briggs,
Charles. A Grammar of the Greek Language.
and Bros., 1855.
Argyle, Aubrey W. An Introductory Grammar of New Testament Greek.
Basil F. C. The Greek Language.
Bauer, Walter; Arndt, William F.; and Gingrich, F. Wilbur. A Greek-
English Lexicon of the New Testament and Other Early Christian
Literature. 2d ed., revised and augmented by F. Wilbur Gingrich
and Frederick W. Danker.
Allen R., and Smyth, Herbert W. Beginner's Greek Book.
American Book Co., 1906.
Blass, Friedrich. Grammar of New Testament Greek. 2d ed., revised and
enlarged. Translated by Henry S.
Blass, Friedrich and Debrunner, Albert. A Greek Grammar of the New
Testament and Other Early Christian Literature. Translated and
revised by Robert W. Funk.
Bruce, Alexander B. "The Synoptic Gospels." In vol. 1 of Expositor's
Greek Testament. Edited by W.
Wm. B. Eerdmans Publishing Co., 1979.
Brugmann, Karl. A Comparative Grammar of the Indo-Germanic Languages.
vol. 4. Translated by R. Seymour Conway and W. H. D. Rouse.
Buttman, Philip. Greek Grammar. Translated by Edward Robinson. New
Chamberlain, William D. An Exegetical Grammar of the Greek New Testament.
B. V., 1976.
Collinge, N. E. "Voice in the Mycenean Verb."
62 (1969-70): 91-95.
Nichols, Lee and Co., 1861.
Curtius, Georg. Elucidations of the Student's Greek Grammar. Translated
by Evelyn Abbott.
_______. The Greek Verb: Its Structure and Development. Translated
by Augustus S. Wilkins and Edwin B.
Dana, Harvey E. and Mantey, Julius R. A Manual Grammar of the Greek New
Davis, William H. Beginner's Grammar of the Greek New Testament. New
Dictionary of New Testament Theology. S.v. "Guard," by C. Brown.
Dictionary of New Testament Theology. S.v. "Prayer," by H. Schonweiss.
Dictionary of New Testament Theology. S.v. "Presuppositions and Theology
in the Greek New Testament," by
Di Pietro, Robert J. Language Structures in Contrast. Rowley: Newbury
House Publishers, 1971.
Erades, P. A. Points of Modern English Syntax: Contributions to English
Studies. Edited by N. J. Robat with a Foreward by R. W.
Funk, Robert W. A Beginning-Intermediate Grammar of Hellenistic Greek.
2d corrected ed., 3 vols.
Geden, Alfred A., and Moulton, William F., eds. A Concordance to the
Greek Testament. 5th ed. revised by Harold K. Moulton.
Gideon, Virtus E. and Vaughan, Curtis. A Greek Grammar of the New
Gignac, Francis T. A Grammar of the Greek Papyri of the Roman and
Byzantine Periods. 2 vols. Milano: Instituto Editoriale
Cisalpino-La Goliardico, 1981.
Gildersleeve, Basil L. "Stahl's Syntax of the Greek Verb." American
Journal of Philology 29(1908):257-79.
Syntax of Classical Greek. 2 pts.
Scribner's Sons, 1965.
Goetchius, Eugene V. The Language of the New
Charles Scribner's Sons, 1965.
Goodwin, William W. Greek Grammar. Revised by Charles B. Gulick.
Gould, Ezra P. A Critical and Exegetical Commentary on the Gospel
According to St. Mark. ICC. Edited by C. A. Briggs, et al.
Green, Samuel. Handbook to the Grammar of the Greek New Testament.
Clarence B. Let's Study Greek.
Edwin. Essays in Biblical Greek.
Hatch, Edwin and Redpath, Henry A. A Concordance to the Septuagint and
the Other Greek Versions of the Old Testament (Including the
Apocryphal Books). 2 vols.
Hatzidakis, Georg N. Einleitung in die Neugriechische Grammatik. In
Bibliothek Indogermanischer Grammatiken. Bd. 5. Bearbeitet von
und Hartel, 1892.
Hendriksen, William. Exposition of the Gospel according to Mark. Grand
Rapids: Baker Book House, 1975.
Householder, Fred W.; Kazazsis, Kostas; and Koutsoudas, Andeas.
"Reference Grammar of Literary Dhimotiki." International Journal
of American Linguistics. 30 (April 1964): 1-188.
Howard, Wilbert F.; Moulton, James H.; and Turner, Nigel. A Grammar of
New Testament Greek. 4 vols.
Huddilston, John H. Essentials of New Testament
Macmillan Co., 1915.
Huther, J. Ed. Critical and Exegetical Handbook to the General Epistles
of James, Peter, John, and Jude. In Meyer's Commentary on the
New Testament. 3d ed. vol. 10. Translated by Paton J. Gloag,
et al. Funk and Wagnalls,
1884; reprint ed.
Jannaris, Anthony N. An Historical Greek
Eric G. New Testament Greek: An Introductory Grammar.
Jelf, William E. A Grammar of the Greek Language. 2d ed. 2 vols.
Johnstone, Robert. Lectures Exegetical and Practical on the Epistle of
James. Oliphant Anderson and Ferrier, 1871; reprint ed.,
Joint Association of Classical Teachers Greek Course. Reading Greek:
Grammar, Vocabulary and Exercises.
University Press, 1978.
Kennedy, H. A. A. Sources of New Testament Greek: The Influence of the
Septuagint on the Vocabulary of
the New Testament.
T. and T. Clark, 1895.
Kuhner, Raphael. Grammar of the Greek Language, for the Use of High
Schools and Colleges. Translated by Bela B. Edwards and Samuel
G. W. H. A Patristic Greek Lexicon.
Lancelot, Claude. A New Method of Learning the Greek Tongue. 2 vols.
Translated by Thomas Nugent.
Law, Robert. The Tests of Life: A Study of the First Epistle of St.
John. 3d ed.
Leddusire, Frank E. B. "A Comparative Study of the Middle Voice in Koine
Greek and Reflexive Verbs in Old Russian through Case Grammar
Lenski, R. C. H. The Interpretation of the Epistles of St. Peter, St.
John, and St. Jude.
The Interpretation of St. Mark's Gospel.
burg Publishing House, 1961.
Liddell, Henry and Scott, Robert, compilers. A Greek-English Lexicon.
Revised and augmented throughout by Henry Jonas with the assis-
Roderick McKenzie et al., with a supplement.
At the Clarendon Press, 1968.
Machen, J. Gresham. New Testament Greek for Beginners.
Macmillan Co., 1951.
Mandilaras, Basil G. The Verb in the Greek
Hellenic Ministry of Culture and Sciences, 1973.
Joseph B. The Epistle of St. James. 3d ed.
Mayser, Edwin. Grammatik der griechischen Papyri aus der Ptolemderzeit.
2 bds. 3 teils.
Intosh, Roy S. Greek Forms and Syntax.
Meyer, Heinrich A. W. Critical and Exegetical Hand-book to the Gospels
of Mark and Luke, vol. 2. Translated by Robert E. Wallis. In
Meyer's Commentary on the New Testament. Revised and edited by
William P. Dickson.
Meyer-Myklestad, Johannes. An Advanced English Grammar for Students and
Milligan, George and Moulton, James H. The Vocabulary of the Greek
Testament Illustrated from the Papyri and Other Non-Literary
Misra, Satya S. A Comparative Grammar of Sanskrit, Greek, and Hittite.
by Suniti K. Chatterji.
Moule, C. F. D. An Idiom Book of New Testament Greek. 2d ed.
Moulton, James H. An Introduction to the Study of New Testament Greek.
5th ed. Revised by Henry G. Meecham.
Henry P. The Elements of New Testament Greek.
University Press, 1918.
A Short Syntax of New Testament Greek.
University Press, 1912.
Oesterley, W. E. "The General Epistle of James." In vol. 4 of Exposi-
tor's Greek Testament. Edited by W. Robertson Nicoll. Reprinted
Leonard R. The Greek Language.
Plummer, Alfred. An Exegetical Commentary on the Gospel according to
St. Matthew. 3d ed.
Rahlfs, Alfred, ed. Septuaginta.
Robertson, Archibald T. A Grammar of the Greek New Testament in the
Light of Historical Research.
Ropes, James H. A Critical and Exegetical Commentary on the Epistle of
St. James. ICC. Edited by C.
A. Briggs, et al.
Charles Scribner's Sons, 1916.
Ruck, Carl A. P. Ancient Greek: A New Approach.
Anatol F. The Greek Language in Its Evolution.
Shilleto, Richard. "On Greek Deponent Verbs with Aorist in qhn." The
Journal of Philology. 7(1875):148-51.
Simcox, William H. The Language of the New Testament. Reprint ed.
Gustave. A Greek Grammar, Accidence.
Charles R. Tongues in Biblical Perspective.
Smith, David. "The Epistles of John." In vol. 5 of Expositor's Greek
Testament. Edited by W. Robertson Nicoll. Reprint ed. Grand
Rapids: Wm. B. Eerdmans Publishing Co., 1979.
Smyth, Herbert W. Greek Grammar. Revised by Gordon M. Messing.
Sonnenschein, Edward A. A Greek
Trubner and Co., 1914.
Sophocles, Evangelinus A. Greek Lexicon of the Roman and Byzantine
Periods. 2 vols.
Stahl, Johann M. Kritisch-historische Syntax des griechischen Verbums
Steyer, Gottfried. Satzlehredes
F. Ullman KG, 1972.
Story, Cullen T. K. and Story, J. Lyle. Greek to Me: Learning New
Testament Greek through Memory
Harper and Row, 1979.
Herbert. Introduction to the History of Language.
Longmans, Green, and Co., 1891.
Svartvik, Jan. On Voice in the English Verb.
Mouton and Co., 1966.
Tasker, R. V. G. The General Epistle of James. TNTC. Edited by R. V. G.
Vincent. The Gospel according to St. Mark. 2d ed.
Macmillan and Co., 1966.
Thackeray, Henry St. J. A Grammar of the Old Testament in Greek accord-
to the Septuagint.
Thayer, Joseph. Greek-English Lexicon of the New Testament. Grand
Rapids: Zondervan Publishing House, 1975.
Theological Dictionary of the New Testament. S.v. "ai]te<w," by Gustav
F. E. A Syntax of Attic Greek.
Thompson, John. A Greek Grammar, Accidence and Syntax for Schools and
Thrax, Dionysius. Grammatici Graeci. vol. 1. Lipsiae: In Aedibus
B. G. Teubneri,
1838; reprint ed.,
Verlagsbuchhandlung Hildescheinz, 1965.
Nigel. Grammatical Insights into the New Testament.
T. and T. Clark, 1965.
Valpy, Richard. The Elements of Greek Grammar.
Veitch, William. Greek Verbs, Irregular and
Clarendon Press, 1879.
Walther, James A. New Testament Greek Workbook: An Inductive Study of
the Complete Text of the Gospel
Warburton, Irene P. "On the Verb in Modern Greek." Ph.D. dissertation,
John W. The Elements of New Testament Greek.
sity Press, 1965.
Kegan, Paul, Trench, Trubner and Co., 1919.
Brooke F. The Epistle to the Hebrews.
Winer, George B. A Grammar of the Idiom of the New Testament. 7th ed.
Enlarged and improved by Gottlieb Lunemann.
Zerwick, Maximilian. Biblical Greek. Adapted from the 4th Latin ed. by
send any errors discovered to: Ted