Syntactic structures

When studying language, we only have discourses as a starting point. The grammar of the language is not visible in itself. The syntactic structures are not visible for observation, but there are good indications that the system of rules that constitutes the competence of the speakers is formed by syntactic structures. Let’s exemplify. A first evidence is in phrases like:

• Pedro found the password for the credit card he had lost.

The given example is a syntactic ambiguity. After all, Pedro lost his password or card? We can resolve the confusion with the use of parentheses. Watch:

• Pedro found the password (for the credit card he had lost).
• Pedro found (the credit card password) that he had lost.

The use of parentheses generates two valid meanings for the sentence. This indicates that the way we group the items of the sentence influences its meaning. Another evidence is in the case category, present in several languages. In Portuguese, we have an example of this category in personal pronouns. Note the following sentences:

• I handed the book to Peter.
• Pedro received the book from  me .

In both sentences, the same person is sometimes designated by the word  I , sometimes by the word  me . This is because personal pronouns have different forms depending on the case. And what would be the considered difference that justifies the use of different forms of the pronoun? It is not a difference that can be explained by semantic reasons such as variations in gender or number.

The use of different forms of the pronoun in the examples is due to the fact that the pronoun is being used in different syntactic contexts. In the first sentence, the pronoun plays a role in the syntactic structure of the sentence, and in the second, it plays another. Classical Latin is a language that makes extensive use of case inflections. Nouns, in Classical Latin, have six different forms, one for each case, or in other words, one for each role played in the sentence structure.

Faced with evidence like this, we are led to believe that speakers actually segment sentences into parts, relating each part to a syntactic function, both when generating sentences and when interpreting them. Speakers are intuitively aware of these parts, so much so that in languages ​​like Latin, they employ different forms depending on the syntactic role played by the word in the sentence. These segments are organized into a syntactic structure. It is these structures that we will deal with hereafter.

Formal units of syntax

At the syntactic level of analysis we find three formal units: period , phrase and phrase . We can say that the model is concentric. The period is the upper unit and is made up of phrases. These, in turn, are composed of phrases that are the lower units of the syntactic level. Note the example:

• (Niemeyer designed the buildings) and (Lúcio Costa created the urban project.)

In the example, we have a period composed of two phrases:

• Niemeyer designed the buildings.
• Lúcio Costa created the urban project.

Each sentence in the example is broken down into phrases.

Niemeyer designed the buildings.

• Subject :  Niemeyer .
• Verb phrase :  projected .
• Direct object :  the buildings .

Lúcio Costa created the urban project.

• Subject :  Lucio Costa
• verb phrase :  created
• Direct object :  the urban project.

Unit Fittings

A deeper analysis, however, shows us that thanks to the fitting resource, we can find periods contained in periods, phrases contained in phrases and phrases contained in phrases, in addition to periods contained in phrases, periods contained in phrases and phrases contained in phrases . See examples.

Period contained in period:

• [I called him to the meeting], but [(he claimed a full schedule) and (did not guarantee attendance.)]

Period contained in sentence:

• Everyone expects [(he reconsiders) and (forgive his friend.)]

Period contained in phrase:

• We have {the hope that [(justice is slow) but (does not fail.)]}

Sentence contained in sentence:

• [I guarantee (he will come to the meeting.)]

Phrase contained in phrase:

• I was obsessed [with the idea that (that I could achieve the goal.)]

Phrase contained in phrase:

• She is [daughter of (Cláudia and Márcio.)]

Apparently, the various possibilities of fitting that syntax allows us disturb the idea that period, sentence and syntagm are organized in a concentric model. The contradiction is apparent and we will explain why.

Periods are made up of sentences and sentences are made up of phrases. This principle is our starting point, valid for all statements, excluding only the cases in which it fits. In fitting, a syntactic formal unit appears in an unexpected position, for example: a period takes the place of a sentence or a sentence takes the place of a phrase. But when the higher-level unit occupies the position of a lower-level unit, the higher-level unit assumes the syntactic profile of the lower-level unit.

Taken as a whole, when a sentence is embedded in another, it becomes a syntagm of the mother sentence. It simultaneously becomes a syntagma of the parent phrase and the phrase if considered in isolation. The docking feature makes the docked item take on dual functionality. In docking, a segment can be both a lower-level unit in the context of the parent unit and a higher-level unit if considered in isolation.

Delimitation of the formal units of syntax

It is not easy to rigorously define the formal units of syntax (period, phrase and phrase), given the complexity and richness of the grammatical structure. This statement contrasts with the ease with which speakers intuitively grasp these concepts.

Language users quickly understand what we mean by the term  phrase , if we introduce the concept through examples, if we make approximations or if we appeal to the speakers’ intuition. In fact, Traditional Grammar has always used these devices to present the concept of a sentence. But when we look for a formal definition for the phrase, the situation becomes considerably complicated.

Likewise, speakers have a clear intuition of what a sentence is, as they can easily segment speeches into periods. They do this when, in written speech, they use capital letters and punctuation marks to delimit the beginning and end of sentences. In oral speech, they delimit sentences using pauses and intonation modulations. However, if we want a rigorous definition of period, we need to get around some difficulties.

As for phrases, it is difficult to say to what extent speakers have a strong intuition of these formal units, as there are few contexts in which the user is aware of them, except in the sphere of grammatical studies.

Next, we will present relevant considerations for the delimitation of the formal units of syntax.

Start and end delimiters

An immediate finding about periods is that in written speech, periods begin with a capital letter and end with a final punctuation mark. These findings, valid for most periods, lead us to the question: are these characteristics relevant to the definition of the period or are they a consequence of the fact that the utterance is a period? Do we capitalize the beginning of a period because we recognize it as a period or to establish it as a period? Does the initial capital just reinforce information contained in other elements that we are facing a period? Is this a redundant sign that, if removed, does not harm the speaker‘s understanding, who can easily locate himself within the context and delimit the period, even without the presence of a beginning graphic sign?

What about final punctuation marks? Well, they are decisive to define the type of sentence that composes the period, mainly, to distinguish between interrogative and non-interrogative sentence. If we change the period to the question mark, we will see that in both cases we still have a sentence, although of a different nature from one another.

The findings we made about written discourse can be transposed to oral discourse. When pronounced, periods follow definite intonation patterns. Is this an important characteristic for the concept of a period, or, the intonation reinforces the information that the segment is a period. It is worth remembering that the intonation patterns also serve to distinguish between types of sentences that make up the sentence, in particular, to differentiate the interrogative ones from the non-interrogative ones.

We asked these initial questions to address the problem present in the following two statements:

• I came, I saw, I won.
• I came. Saw. I won.

In the first statement, in which we have only one period, there is only one period. In the second statement, with the segments delimited by three full stops, we have three periods. From the example, we see that in some cases the use of start and end delimiters is crucial for establishing the segment as period. In the given example, without the delimiters we cannot anticipate the sender’s proposal for segmenting the utterance into periods.

The conclusion is that, in some cases, the delimitation of the periods of an utterance is established by the will of the sender, depending on the degrees of freedom that the utterance offers. If the sender considers that the three sentences in the example form a concatenated set, he organizes them into a larger grouping, the period, and makes this clear to the receiver through the start and end flags.

On the other hand, if the sender understands that the phrases of the utterance do not form a concatenated group, it makes a different distribution of the beginning and end delimiters.

Not all utterances offer more than one segmentation possibility. Analyze the following example:

• he tried hard but didn’t reach the minimum grade

We present the utterance without the beginning and end delimiters, but it will still be possible to delimit the period and its phrases, because in this case there is only one possibility of segmentation that produces acceptable utterances.

• He tried hard, but he didn’t hit the minimum mark.

In short: start and end delimiters cooperate with other elements of syntactic structure to establish the formal units of syntax. In some cases, they are redundant and, if subtracted, it will still be possible to discern which is the correct segmentation of the utterance. In other cases, the use of delimiters is crucial to determine the targeting proposed by the issuer. If they are subtracted, an uncertainty arises that is not dissipated by analyzing other elements of the syntactic structure.

full statement

A basic characteristic of formal syntactic units is that they are complete utterances. This means that the information contained therein forms a consistent whole and that the absence of elements necessary for understanding is not identified. When we hear or read a well-formed sentence, for example, our understanding does not detect absences that cause us strangeness or difficulties in decoding. But it should be remembered that the completeness of the formal units takes place in the grammatical layer of analysis and not in the contextual layer. The grammatically complete utterance and the contextually complete utterance are different. For example:

• I go.
• I will go with you to…

The first utterance can be considered grammatically complete, but it may be insufficient to satisfy the communication needs in a specific context. The listener may ask: Where are you going? However, if the listener is immersed in a context where he knows where the sender is going, communication takes place satisfactorily.

The second utterance is grammatically incomplete, but perhaps it is enough for the receiver to grasp the meaning of the message. Grammatically incomplete because in Portuguese a noun phrase is required after the preposition. The absence of this phrase generates strangeness to the receiver. On the other hand, if the receiver knows from the context where the sender is going, the communication happens without problems.

self-contained utterance

The parts of an utterance are interconnected by a network of syntactic relationships. One of the characteristics of syntactic formal units is that the relationships are resolved within the limits of the unit, that is, the units are self-contained from the syntactic point of view and do not depend on external elements to be comprehensible.

syntactically well-formed utterance

To say that an utterance is well-formed is to refer to a series of conditions that must be satisfied for well-formedness to be confirmed. Undoubtedly, there is an enormity of grammatical rules that must be obeyed so that an utterance can be called well-formed. But for our delimitation of formal units, the most important condition is the organization of syntactic elements according to valid models of the grammar of the language. In other words: in the formal unit, we can recognize segments of a lower level, but which belong to the syntactic level of analysis, organized according to one of the valid models of the grammar of the language. There should be no lack of elements, no excess, no conflict, no problems in the order in which they are distributed, etc.

Traditionally, effort has always been spent on determining what would be the essential syntactic constituents of the sentence. The effort was in vain, as there are probably no such constituents. Yes, there are several combinations of syntactic constituents that are recognized as a sentence, however, in these combinations we were unable to extract a universal regularity or essential component. Perhaps the only universal characteristic that can be attributed to sentences is that they are constituted by a grammatical combination of lower-level syntactic constituents, among which we highlight: subject, verb phrase, adjective phrase, direct object, indirect object and adverbial phrase.

minimal statement

One of the conditions for an utterance to be a formal syntactic unit is that it is not possible to divide it into two other segments in which the two in turn are also complete, self-contained and syntactically well formed.

Interdependence between phrase and phrase

One of the difficulties for the definition of phrase and phrase is the interdependence of the two concepts, which makes it difficult to define one before the other. We only recognize a well-formed, or acceptable, sentence to the extent that we identify the phrases that compose it. On the other hand, phrases are only understandable insofar as they are seen as constituents of the sentence.

This problem is millenary, so much so that in Traditional Grammar the most typical model of sentence is defined as a statement that presents a subject and a predicate, or at least a predicate.

Ellipses

Note the following period:

• I was lucky and you weren’t.

In this example, we have two ellipses, that is, the predictable absence of some item in the sentence. The first ellipse is the absence of the pronoun  I  in the sentence  I was lucky . The second ellipse is the absence of was lucky  in the phrase  you, no. The example period can be paraphrased as follows:

• I  was lucky and you  were unlucky .

The ellipse of the pronoun  I  is possible because the subject of the action is implied in the inflection of the verb. Tive  is the first person singular inflection of the verb ter , which is enough to determine the subject of the action as  eu . The second sentence of the example period, taken by itself, is unacceptable.

* You don’t.

However, the ellipsis is allowed in this case because the elided items were cited in the previous sentence. This is a particular case of an ellipse, known as a zeugma.

The ellipse is a feature widely used in all linguistic variants and constitutes a challenge for syntactic theory. Syntactic theory is based on the analysis of complete sentences. The ellipsis disturbs the syntactic analysis because there is no way to guarantee, except by speculation, how the elision of items in the utterance took place

In our analysis, we will focus on complete utterances. Let’s consider that the ellipse is an operation after the syntactic structure of the sentence. We will treat the ellipse as an operation beyond the syntactic level of analysis.