empirical approach muddled in an infinitude of messy
details.2 His approach, which revolutionized the
linguistic world, was to isolate a few, simple, syntactic
____________________
1Leech, Semantics, pp. 32-33. It is
interesting that Philip Pettit has attempted to show the dependence
of Chomsky on Saussure (The Concept of Structuralism: A
Critical Analysis [Berkeley: University of California
Press, 1977], pp. 1-28). Lockwood, Introduction to
Stratificational Linguistics, p. 263. Some easy beginner
texts which introduce the concepts of TG (transformational
grammar) are: Herndon, A Survey of Modern Grammars; Brown
and Miller, Syntax: A Linguistic Introduction to Sentence
Structure; and Liles, An Introductory Transformational
Grammar (Englewood Cliffs: Prentice-Hall, Inc., 1971).
Radford's work (Transformational Syntax [1981]) is more
advanced and up to date.
2Noam Chomsky, Cartesian Linguisitics: A
Chapter in the History of Rationalist Thought (New York:
Harper & Row, Publishing, 1966), pp. 3-30.
rules which could generate all possible sentences. This
is why it is called "generative grammar" rather than a
"descriptive grammar."1 His system, like the
structuralist's, still focused on the syntactic component
of language as foundational.2 Others have more recently
opted for a semantic base and consequently called it
"generative-semantics."3 Chomsky's system attempts to
explain: syntactically "synonymous" sentences which have
different meanings ("John is easy to please"; "John is
eager to please"); sentences which are syntactically
ambiguous ("Visiting relatives can be tiresome," or
"Flying planes can be dangerous"); and sentences which are
syntactically different yet "synonymous" ("Brent painted
the picture"; or "The picture was painted by Brent").4
Two sets of distinctions are important in
transformational grammar. First, a distinction is made
____________________
1John Lyons, Noam Chomsky (New York: Viking Press,
1970), p. 9; Herndon, A Survey of Modern Grammars, pp.
118-20; Liles, An Introductory Transformational Grammar
p. 133; and Radford, Transformational Syntax, pp. 19-20.
2Noam Chomsky, Syntactic Structures (The
Hague: Mouton, 1957), pp. 11-17; also, later, his Studies on
Semantics in Generative Grammar (The Hague: Mouton, 1972),
pp. 11-14.
3Chafe, Meaning and the Structure of Language; and
Ray S. Jackendoff, Semantic Interpertation in Generative
Grammar (Cambridge, MS: The MIT Press, 1972). Also vid.
the works of George Lakoff and James McCawley.
4Herndon, A Survey of Modern Grammars, p. 121; and
Liles, An Introduction to Linguistics, p. 133.
between competence (fluent native speaker's knowledge of
his language) and performance (that which he actually
speaks).1 Transformationalists argue that those linguists
who have analyzed a corpus are merely studying
performance, rather than speaker competence, which should
be the object of language study.2 Chomsky then attempts
to describe competence through a series of syntactic rules
by which the mind generates sentences. Second, the
distinction between deep and surface structure, with
transformations mediating between these two levels, has
been a major contribution of transformational grammar.
"Deep structure refers to the basic syntactic pattern in
which a meaning is expressed, while surface structure
refers to the particular form in which a meaning is
expressed in a text."3 This was another clear move away
from empiricism. Thus, two sentences, such as "Joy was
hit by the ball" and "The ball hit Joy," were now able to
be compared for the deep grammar similarity, even though
their surface level is syntactically discordant. The deep
grammar is described by a series of phrase structure rules
which are the same for both of these sentences. Next a
____________________
1Noam Chomsky, Aspects of the Theory of Syntax
(Cambridge, Mass: The MIT Press, 1965), p. 4; cf. also
Radford, Transformational Syntax, p. 2.
2Liles, An Introductory Transformational Grammar,
p. 79.
3Louw, Semantics of New Testament Greek, p.
73.
series of transformational rules maps the common deep
structure onto the different surface structures (one
active, one passive). A series of phonological rules
takes the results of the transformations and projects them
into exact speech sounds. Therefore, there are three
levels of rules to which a fourth must be added: deep
phrase structure rules, transformations, lexical rules
(which plug in the appropriate words choices), and
phonological rules.1
An example may prove beneficial at this point.
One type of sentence may have the deep phrase structure
generating rules:
S --- NP + VP (Sentence consists of a NP and VP)2
NP--- N (Noun Phrase consists of a Noun)
VP--- V + NP (Verb Phrase consists of a V + NP)
NP--- Art. + N (NP consists of an article + N)
These phrase structure rules would generate any of the
following sentences and many more (any of the type N + V +
Art. + N):
Dawn cut the flowers.
Dogs ate the fish.
Children threw the ball.
Firemen extinguished the blaze.
If a passive transformation is applied to the
____________________
1Liles, An Introduction to Linguistics, pp. 72,
168, has helpful charts of this process. Cf. also Herndon,
A Survey of Modern Grammars, p. 125; and Radford,
Transformational Syntax, pp. 15-16.
2The arrow means "consists of" or "has the constituents of."
phrase structure, one arrives at a different surface
structure, but one which is derived from the same deep
phrase structure. The passive transformation does the
following to the initial phrase structure: N1 + V + Art.
+ N2 === Art. + N2 + Aux. + V + by + N1 (where aux. =
verbal auxiliary). Thus, this transformation accounts for
all surface structures which are "synonymous" with the
original sentences in a deep structure sense, but very
different in the surface structure. This transformation
results in:
Dawn cut the flowers ===> The flowers were cut by
Dawn.
Dogs ate the fish ===> The fish were eaten by dogs.
Children threw the ball ===> The ball was thrown by
children.
Firemen extinguished the blaze ===> The blaze was
extinguished by firemen.
Other transformations explicitly explain the relationship
between statements and questions ("He is tall" and "Is he
tall?"); indirect object transformations ("Kathy gave him
a shot" becomes "Kathy gave a shot to him"); adverbial
movement transformations ("Yesterday I saw Rik" becomes "I
saw Rik yesterday"); compounding, deletion and pro forms
("Skip was eager and Skip was industrious" becomes "Skip
was eager and industrious" or "Skip was eager and he was
industrious"); as well as relative constructions ("He is
building a boat" and "The boat is large" become "The boat
that he is building is large").1 It is clear, because of
our native competence (fluency) in English, that these
sentences are related and many of them would be considered
"synonymous" in normal speech. The exegetical
ramifications are astounding, but will not be pursued in
this study other than to say that these examples
demonstrate that one must be extremely careful about
making eisogetical remarks on the basis of surface grammar
variations with deep structure "synonymity." It is
possible that the writer was not attempting to make any
crucial point by his choice of a passive rather than an
active mode. Furthermore, the transformational idea holds
rich possibilities for Hebrew syntactic parallelism. This
writer has observed bi-cola which are syntactically
non-matching, according to O'Connor's system, but which,
with a simple transformation, match perfectly (viz. Prov
10:1).2
____________________
1Herndon, A Survey of Modern Grammars, pp. 207-43;
and Liles, An Introductory Transformational Grammar, pp.
43-101.
2It is interesting that William Mouser's recent
book on Proverbs proposes a similar idea--only somewhat
non-scientifically specified--using semantic
transformations to allow for a better fit between the
bi-cola (Walking in Wisdom: Studying in the Proverbs of
Solomon [Downers Grove, IL: InterVarsity Press, 1983], pp.
35-52). One must be extremely cautious, however, of this
approach, as it may force the proverbs to fit equivalence
patterns without seeing the variety and differences in
meaning intended. Thus this methodology, while having
Several objections have been raised to
transformational grammar.1 Robinson is correct when he
critiques TG for treating only sentences, which are only a
single level of language.2 He further objects that
language is more than a series of rules and that such a
rule-oriented approach chases meaning out of language.3
Chafe accuses Chomsky of "syntacticism."4 How does one
handle sentences which are ungrammatical, but are spoken
nevertheless? Is not TG a return to prescriptivism? How
does TG handle the metaphors, irony, and perlocutionary
acts (the effect on the hearer) of language?5 It is
ironic that Robinson correctly accuses the "rationalistic"
approach of Chomsky as empiricism revisited.6 Though
____________________
possibilities, needs further development along
scientifically semantic lines.
1Robinson's book, The New Grammarians' Funeral, is
perhaps the most acrid, written from an intuitive/
impressionistic approach. More linguistically satisfying
is R. A. Hudson, Arguments for a Non-transformational
Grammar (Chicago: The University of Chicago Press, 1976).
2Robinson, The New Grammarians' Funeral, pp. 36,
45.
3Ibid., pp. 21, 87.
4Chafe, Meaning and the Structure of Language, p.
60.
5Robinson, The New Grammarians' Funeral, pp.
10, 40, 47.
6Ibid., p. 104. It seems Chomsky desires to
project empirical evidence back into the mind via his
rules. How can one validate or falsify such mentalistic
suggestions?
"Sincerity admires Rebekah" may be semantically ill-formed
by the "rules"--because "admires" needs an animate human
subject--yet, in poetry, such a sentence may be
well-formed.1 The problem is not with grammar per se, but
with attempting to reduce language to mere grammar.2
While the idea of transformations is very helpful in
relating sentences, problems arise if one views
transformations as producing equivalent or exactly
synonymous sentences. Such an approach would deny the
passive a reason for existence, portray repetitions as
jejune redundancies, and tend to de-emphasize the
importance of surface structure selectional options.3
Hence, transformations may result in a leveling of the
meaning of the text via a syntactic reductionism which
manifests an "X is really Y syndrome."4 It should be
noted that tagmemics clearly distinguishes surface
____________________
1Freeman, Linguistics and Literary Style, pp.
182-83.
2Robinson, The New Grammarians' Funeral, p.
60.
3Ibid., pp. 119, 125; M. K. Brame, Essays Toward
Realistic Syntax (Seattle: Noit Amrofer, 1979), p. 14;
Daniel Gulstad, "Are Transformations Really Necessary?" in
Papers from the 1977 Mid-America Linguistics Conference,
ed. Donald Lance and Daniel Gulstad (Columbia, MO:
University of Missouri, 1978), p. 203; and Rolv Blakar,
"Language as a Means of Social Power: Theoretical-
Empirical Explorations of Language and Language Use as
Embedded in a Social Matrix," in Pragmalinguistics: Theory
and Practice, ed. Jacob Mey (The Hague: Mouton, 1979), p.
152.
4Chafe, Meaning and Structure of Language, p.
86.
structures, while at the same time--through an embedded
case grammar--accounts for deep structure regularities.
Robinson champions Occam's razor. He misses the point,
however, that Chomsky's categories do have value if left
on a grammatical level.1 Thus, TG must remain a grammar
rather than a total theory of language. Its formulae will
bear some correspondence to the filler box of the tagmemic
formulae. Tagmemics remedies many of the above problems
and will, therefore, be adopted in this study, although
some will obviously disparage the use of a non-main-stream
grammar. However, the advantages of tagmemics out-weigh
this criticism and the similarity of tagmemics to TG makes
it easily learned by those familiar with TG.
Other Recent Grammars
For several reasons, it is desirable to survey, in
a very brief fashion other approaches to linguistics:
(1) to suggest other directions which this study may have
taken; (2) to allow for a comparison with the tagmemic
system adopted here; (3) to help sensitize the reader to
aspects of language which tagmemics has ignored; and (4)
in the spirit of eclecticism, to suggest factors which may
be beneficially incorporated into a tagmemic analysis.
Two grammars (stratificational and daughter-dependency)
will be mentioned. Lastly, and with great promise, the
____________________
1Robinson, A New Grammarians' Funeral, pp.
x, 165.
recent developments in pragmalinguistics will be broached.
Stratificational Grammar
Stratificational linguistics was developed in the
late sixties by Sydney Lamb1 and H. A. Gleason.2 A
work by David Lockwood provides a helpful introduction to this
theory.3 Walter Bodine, at a recent colloquium, has
alluded to some work which is presently taking place at
Dallas Seminary applying this theory to Hebrew.4 Its
diagrams specify relationships, treating units only as
input or output items. Like tagmemics, stratificational
linguistics allows for relationships on the various
levels, developing a "tactic" system for each level
(phonotactics, morphotactics [syntax], lexotactics,
semotactics).5 Once the diagram is entered it is totally
____________________
1Sydney Lamb, Outline of Stratificational
Grammar (Washington, DC: Georgetown University Press, 1966).
2Henry A. Gleason, "The Organization of Language:
A Stratificational View," Monograph Series on Languages and
Linguistics, no. 17, ed. C. I. J. M. Stuart 17 (Washington,
DC: University Institute of Languages and Linguistics,
1964).
3David G. Lockwood, Introduction to
Stratificational Linguisitics (New York: Harcourt, Brace,
Jovanovich, Inc., 1972); and a bibliograhy by Ilah Fleming,
"Stratificational Theory: An Annotated Bibliograhy,"
Journal of English Linguistics 3 (1969):37-65.
4Walter Bodine, "Linguistics, Semitics, and
Biblical Hebrew," in Society of Biblical Literature 1982
Seminar Papers 21 (Chico, CA: Scholars Press, 1982):31-37.
5Lockwood, Introduction to Stratificational
Linguistics, p. 26.
relational, as opposed to a more constituent oriented
approach such as tagmemics or TG. Stratificational
linguistics handles the following three types of
relationships: (1) conjunction/disjunction; (2) ordered/
unordered; and (3) downward/upward. It uses a series of
"and" and "or" gates which are similar to systems
engineering models. The "and" gate calls for both items
to be present and the "or" gate requires that a selection
be made, with one item being chosen. Thus, for example
the sentence "Perry/Elaine/Dave sees Donna" would be
diagrammed as follows (the convex triangular shapes are
"and gates" and the sideways parenthesis is an "or gate"):
"Perry/Elaine/Dave sees Donna" [OUTPUT]
Perry Dave see s Donna [INPUT]
Elaine
Notice that the input is words and the output is a
sentence.1 One can see that this system is also, like
tagmemics, hierarchical in nature, showing relationships
____________________
1Ibid., p. 35.
from the morpheme up to discourse level. While
stratificationalists have noted the similarities with
tagmemics,1 they have ignored at least two fundamental
differences which are: (1) the fact that constituents or
units are crucial to the theoretical underpinnings of
tagmemics;2 and (2) tagmemics attempts not only to note
the relationships between units, but also--and very
important for this study--to specify the exact nature of
what those relationships are.3 Thus, the stress on
relationships is very beneficial, but the need for
constituents at each level and the exact specification of
relationship types (via case grammar) will make it
desirable to pursue a tagmemic approach.
Relational Grammars
Another more recent set of approaches has been
through relational, dependency and daughter-dependency
grammars. These models develop the European dependency-
____________________
1Ibid., pp. 255-57.
2Cf. Pike's comments in "On Describing
Languages," in The Scope of American Linguistics, ed. Robert
Austerlitz (Lisse: The Peter De Ridder Press, 1975), pp. 13, 24; and
his discussions with S. Lamb in Report of the Twenty-Second
Annual Round Table Meeting on Linguistics and Language
Studies, ed. Richard J. O'Brien (Washington, DC:
Georgetown University Press, 1971), p. 158.
3This writer is also aware of attempts to embed
case grammar into this system. Lockwood, Introduction to
Stratificational Linguistics, p. 142.
type grammars of Heringer (1970), Vater (1975)1 and Werner
(1975).2 Like stratificational and tagmemic grammars,
dependency grammars separate semantic, syntactic and
phonological levels. Like tagmemics, they are also
constituent-oriented, as opposed to stratificational
grammar.3 While they do diagram constituents
heirarchically (i.e., Nouns and Verbs combining to form
higher level clauses), they eliminate the NP and VP levels
and go right to the V and N constituents.4 Dependency
grammar also does away with the TG concept of deep
structure and transformations.5 The diversity of this
system may be seen in the fact that it monitors
mother-daughter relations (i.e., the relation of higher
level nodes [mothers] to lower level units [daughters]) as
well as sister relations between units on the same level
(cf. relational grammars). Dependency grammar employs
four categories, which are: (1) feature-based rules
(specifying one item as before another [article must
____________________
1Heinz Vater, "Toward a Generative Dependency
Grammar," Lingua 36 (1975):121-45; and Laurie Bauer, "Some
Thoughts on Dependency Grammar," Linguistics 17
(1979):301-15.
2Richard A. Hudson, Arguments for a
Non-transformational Grammar, p. 200.
3Ibid., p. 11.
4Ibid., p. 60.
5Ibid., pp. 1, 14, 131.
precede the noun, for example] or which features come with
a certain item); (2) function-based rules (which order the
three functions, viz., subject, topic, relator);
(3) peripherality-based rules (order the units according
to their peripheral nature); and (4) dependency-based
rules (which attempt to describe the relationship between
sisters).µ1µ An illustration of this approach may help.
From the following diagram one will be able to see that
this system stresses horizontal relationships among
sisters.
+sentence
+interrogative
etc.
Dostları ilə paylaş: |