to be incorporated into reading materials.
2.6) Some Sources of Syntactic Complexity
Linguistics has had, as one of its goals, describing linguistic complexity that is determining which structures are complex. To achieve this objective, linguists tried to develop scientific criteria to measure linguistic complexity, and later, along with psycholinguists, they attempted to figure out how linguistically complex structures affect performance. As Smith (1988) argues it is hard to imagine that linguistic structure has no effect at all on performance and in fact such a view is not generally held (p.217).
Miller and Chomsky (1963), for instance, discuss the processing difficulty of sentences with multiple embedding and Chomsky’s early papers point out that transformations tend to reduce abstract structure, and thereby simplify the task of speaker or hearer (Cited in Miller, 1976).
Nevertheless, linguistic complexity as a component in linguistic processing and language development has been tried to be studied autonomously; that is, independent of performance factors. This attempt was carried out by Chomsky (1981) in his well-known Government and Binding (GB) Theory. Smith (1988) delineates GB Theory as follows:
“It has a modular organization: it is composed of separate, interacting modules that account for lexical, syntactic, interpretive, and phonological properties of linguistic structure. ”
Within the modular framework of GB Theory, Smith (1988) postulates some criteria to determine structural complexity. He proposes that linguistic complexity is a property of individual sentences, and is a result of several factors in combination. He suggests that:
“For a given sentence, a complexity profile is constructed, that indicates its complexity at each level. A sentence may have high complexity of one type and low complexity of another …. Perhaps, the most complex sentences are high in complexity of all kinds. ”
Smith proposes three types of complexity:
1. Surface syntactic complexity
2. Interpretive complexity
3. Systematic complexity
As this proposal can hare some implications for syntactic simplification, a brief explanation of each is given below.
2.6.1) Surface complexity
Surface complexity assesses at surface structure, which is the linguistic level at which all syntactic rules have applied.
The determiners of surface complexity are “Amount”, “Density” and “Ambiguity”.
It refers to the number of linguistic units in a sentence:
it involves words and morphemes. The longer and more morphologically
complex a sentence, the higher its complexity.
It refers to the way linguistic material is distributed in a sentence.
Materials may be distributed quite evenly among the units of a sentence (that is,
NP, PP, etc); or many words may appear in one unit which makes it dense.
Example (1) [Mary wrote NP [a letter] PP [to her family] ]
Example (2) [Mary wrote NP [a letter] PP [about the meeting] ]
(1) has a simple object NP followed by a PP; (2) has a complex object NP, consisting of an NP and a PP. The object of (2) is denser on two counts: it has both more words and more non-terminal node structures within a phrase.
It involves the surface structure interpretation of a sentence. Sentences with more than one bracketing, or category interpretations, are ambiguous.
Example: Visiting relatives can be boring.
The subject noun phrase has two interpretations:
a) (Adj Noun) the relatives who visit
b) (Ving Noun) to visit one’s) relatives
2.6.2) Interpretive Complexity
It arises when the semantic structure that shows scope relations of a sentence is not congruent with its surface structure. It differs from other types of complexity considered here because it involves a comparison between two levels of structure, rather than properties of one level. At least three type of interpretive complexity can be identified: Empty Categories, Discontinuous Constituents, and Semantic Scope.
2.6.3) Systematic Complexity
It deals with the constraints on the rules that produce syntactic structures. Generally speaking, the rules operate without constraints, producing the full range of possible structures. But some words or phrases add to the complexity of a sentence because they constrain the rules. For example, there is a well-known set of words and phrases known as negative polarityitems; they do not appear freely but require a negative of some kind in sentences in which they appear, like left a finger below:
a) I didn’t lift a finger to help her.
*b) I lifted a finger to help her.
The second sentence is odd at best as a literally meant sentence, and ungrammatical on the idiomatic interpretation … to present a proposal for systematic complexity it would be necessary to consider, closely, the form and range of lexical entries.
18.104.22.168) Sentence Length
“The Longer the sentence the more difficult it would be to understand; this has been a major assumption in readability formulas. While this assertion is not counter-intuitive, it does not necessarily prove true all the times. In fact, in some cases, the opposite has been proven. For example, it has been shown that elliptical sentences, though shorter, are difficult because filling in ellipted words demand greater semantico-syntactic and discourse control.
Moreover, length by itself cannot be considered as an independent criterion for sentence difficulty. Rather, its impact should be measured in interaction with other linguistic and extra-linguistic features.
As Anderson and Davison (1988) contend no recent study has focused specifically on the contribution of sentence length per se to comprehension (P.23). Preliminary findings from a study by Davison, Wilson, and Hermon (1985) show that sentence length alone accounts for a very small percentage of variance in the comprehension of text. They suggest that texts with long sentences are comprehended as well as short sentences, except for poor readers.
One evidence demonstrating that long sentences may be easier to understand comes from Davison and Green (1988). They hold that long sentences are usually characterized by connectives which actually facilitate comprehension, especially in a reversible way.
They provide some examples as follows:
1) I moved the switch. The lights went off.
2a) I moved the switch, because the lights went off.
2b) The lights went off, because I moved the switch.
The two sentences in (1) may bear more than one relation to one another. These different interpretations are paraphrased in (2a) and (2b), in which an explicit connective is used.
Therefore, if there is no connective, the reader is not always able to make the correct inference, especially, if it is not clear from the context, which interference, if any, should be made.
The implication for structural simplification would be that sentence length should be observed in context. If a long sentence overrides reader’s short-term memory and makes comprehension difficult, making it shorter or splitting it can be justified. If, on the other hand, sentence length eliminates ambiguity or facilitates comprehension it is better to be left intact.
22.214.171.124) Preposed Clause
It has been shown that preposed clause to be more complex than a similar clause which follows the main verb and its object. Irwin and Pulver (1977) used sentence pairs, like the following, in their research:
A-Because Mexico allowed slavery, many Americans and their slaves moved to Mexico during that time.
B-Many Americans and their slaves moved to Mexico during that time, because Mexico allowed slavery.
It was surprising that the version with the preposed adverbial clause was difficult for the younger subjects. It was predicted that (A) would always be easier that (B), bec
ause the order in which the clauses are mentioned coincides with the general cause-and-effect ordering that is generally preferred. This was the case for older and more skilled readers. This finding has also been supported by another type of reasoning. Yngve (1960), states that words like “the” occur only in phrases with nouns, and precede the noun. This word is a left-branch within a noun phrase and its appearance signals the beginning of a phrase of the NP category. Hence it is stored in working memory while the next constituents are searched for, including the noun. Yngve proposes that for this reason, left branches always require more memory capacity to produce or understand that right branches. Preposed adverbial clauses are left branches, large phrases which mustbe held in working memory until the main clause constituents are found (Bever and Townshend, 1979). Anderson and Davison (1988) remark that the tendency of left branching structure to make a sentence hard to understand is the result of an interaction between the demands on short-term memory caused by left branching structures and a number of other factors.
The implication for syntactic simplification would be: substitute right branching for left branching structures as far as possible.
126.96.36.199) Passive Sentences
The large literature on children’s comprehension of passives (Turner and Rommetveit, 1967; Buldie, 1976; Horgan, 1978; Harris, 1978) has shown that a number of factors reversibility, agent, pragmatic consideration, context, etc-influence the way that such sentences are understood. Slobin (1966) is one of the first researchers who studied how children interpret passives. He gave 6, 8, 10 and 12 year-olds and adults a sentence-picture verification task in which they were presented with a picture and a sentence, and had to say whether the sentences correctly described the picture. He found that, at all ages, reversible passive took longer to verify than non-reversible ones (Cited in Vahdani,1995).
Many experiments, also, have shown (Gluchsberg, Trabasso, and Wald, 1973; Olson and Filby, 1992) passive sentences require less processing time and are more accurately comprehended when the preceding verbal context contains an antecedentfor the passive subjects, which is the topic of the target (Passive) sentences (Vahdani, 1995).
The implication for syntactic simplification would be: when dealing with the passive sentences, pay attention to the kind of passive and its context of use.
188.8.131.52) Relative clause and Embedding
There have been many studies of students’ understanding of relative clauses and a number of theories about why students find them difficult. De Villiers, et al. (1979) identifies two factors that contribute to the structure of relative clauses:
a) Embeddedness: It depends on the position of relative clause within a sentence, or more specifically, which part of the main clause it modifies. If the clause modifies the superficial subject NP, it is center-embedded (CE). If it modifies the object NP, it is right branching (RB).
b) The focus: It refers to the function that the head noun phrase _the noun phrase in the main clause that is modified by the relative clause _ plays in the relative clause (subject, object, or indirect object). Subject-focus relative clauses are often termed subject relative and object-focus are termed object relative. Examples of each type are given below:
1. The cat that bit the dog chased the rat. (CE _ subject-relative)
2. The cat that the dog bit chased the rat. (CE_ object-relative)
3. The cat bit the dog that chased the rat. (RB_ subject-relative)
4. The cat bit the dog that rat chased. (RB_ object-relative)
Sheldon (1974) carried out an influential early study of relative clauses on 3 to 5 year olds and found that neither factors alone could account for the comparative difficulty of different types of relative clause. She proposed the Parallel Function Hypothesis, stating that sentences in which the head NP has the same function (subject or object) in both clauses are easier. Thus (1) and