1. Introduction
We are going to show that linguistic information has a mainly
unobservable component in speech, which cannot be noticed as semantic
information, that is, is nonsemantic and denotes nothing. The
unobservable nonsemantic component is an important characteristic of
speech: It underlies the grammar. The area of study we are going to
present encompasses grammatical facts and, accordingly, could be
pertained to grammar. However, we study not grammatical phenomena
represented in grammatical forms but linguistic information detached
from the form. We don’t turn to a new field of linguistics, but we do
turn to a new level of analysis. This analysis describes linguistic
information on a level deeper than the traditional grammar but less
detailed than the fields relying on semantics. Let us take, for
instance, the sentences
(1) (a) A ball rolls.
(b) The man stands like a kouros.
(c) Two and two make five.
(d) I apologize for my behaviour.
(British National Corpus 2014). All the sentences have the same subject
and verb structure from the point of view of traditional grammar. If we
take into consideration what the denoted things are, we will see the
difference between the sentences: The sentence (c) is false, (d) is a
performative utterance, (b) contains a stative verb unlike (a), which
contains an action verb. Just as the traditional grammar, the
nonsemantic component analysis shows no difference between them. But
this analysis identifies, for example, three types of sentences with a
direct object (they are cited in the section 6), while the traditional
grammar doesn’t distinguish them, and they have no semantic basis for
classification.
The discovered level of grammatical study has no background in existing
branches of linguistics but appears to serve itself as a background, a
foundation of grammar. This level of analysis was not studied earlier.
That’s why the references in the paper don’t reveal its content and
concern only particular cases.
The distinctiveness of this approach is that in describing linguistic
information it is not the word and structure meanings or sentence
significance (construction semantics is further discussed in this
section) that we analyze but a nonsemantic component of them. This
component is semantic information parameters formed by bits of
nonsemantic information. These parameters are not initially embedded in
word meanings but join the meanings used in speech. Unlike the observed
parameters of physical objects, the semantic information parametersdo not represent the meaning’s content, its quantity. They are
just tags that provide a link of one meaning’s parameters to the
parameters of another available meaning and thereby a link between the
meanings. But by linking meaning parameters, sentence grammar is seen as
a unified whole formed on a common ground.
In this paper, we want to show what the level of nonsemantic information
is and give some of the results of its analysis. First, we are going to
show what elements form this level. Then, we will give arguments in
favor of the validity of studying information separate from linguistic
forms and point out various ways of how nonsemantic information
parameters are used. Finally, we will give arguments for the semantic
information parameters existence.
So, this paper suggests and substantiates the hypothesis that semantic
information in speech has a nonsemantic component that distinguishes
speech as language in use from language as list of linguistic tools
(that is, a staticly organized repository of everything that is
necessary for producing speech and can be represented in it).
Nonsemantic by itself, the studied level can be, for the most part,
revealed by means of a disclosure we call linguistic reading. Thus, in
this paper we analyse nonsemantic parameters and their role in forming
sentence information.
To determine the place of this research among the areas of academic
context is a difficult task. We inevitably deal with phenomena being
under study in other areas. However, none of those areas provides a
theory that might be a starting point for this research and none
contains a combined research base on which it could rely. The best way
is, in our view, to show differences of methods and objects of study
between our research and other methods whose object we deal with. To do
that, we must clarify the following notions: Sign and its relation to
meaning in speech; meaning vs. information.
There are different approaches on sign and its relation to meaning in
speech. Wallace Chafe (2018) considers the relation between sign and
meaning, first of all, for the speaker from a whole thought to overt
phonology via a semantic structure: “… organizing thoughts and
pairing them with sounds defines what language is (27). One can validly
object that this figure [thoughts → sounds] ignores a listener’s
task of moving from sounds back to thoughts, but understanding language
depends on having something to understand in the first place” (28).
“The process [“verbalization”] is set in motion by a thought”
(30). The next stage is semantic structuring of thought, then it is a
syntactic structure, which is symbolized by an abstract phonological
structure, then it is overt phonology, “which then provides the input
to sound” (30). André Martinet (1967) suggests the following
segmentation of language: “La première articulation du langage
est celle selon laquelle tout fait d’expérience à transmettre, tout
besoin qu’on désire faire connaître à autrui s’analysent en une suite
d’unités douées chacune d’une forme vocale et d’un sens” (13) [The
first articulation of language is that according to which any experience
to be transmitted, any need that one wishes to make known to others, are
analyzed in a series of units each endowed with a vocal form and a
sense]. The “monème” is a minimal sign which «ne saurait être
analysé en une succession de signes” (15) [cannot be analyzed in a
sequence of signs]. Thus, André Martinet talks about a sequence of
units, so that the sequence itself as well as its smaller components
have a sound form and a sense. Ray Jackendoff (2010) considers the
interface between levels of mental representation: “In the Parallel
Architecture, the combinatorial properties of meaning are not derived
from syntax; they arise from autonomous conceptual combinatoriality.
… Part of the mapping [between meaning and sound (in either
direction)] is provided by the words, which are interface rules
between small-scale pieces of meaning and sound. The remaining part of
the mapping is the encoding of the semantic relations among the
words: the function-argument, head-modifier, and binding relations”
(20). So, R. Jackendoff sees two modes of mapping between meaning and
sound: pieces of meaning and sound in words, and semantic relations and
sound above the words. To express our view on sign and its relation to
meaning in speech, let us cite Ferdinand de Saussure’s well-known
statement: “les signes acoustiques ne disposent que de la ligne du
temps; leurs éléments se présentent l’un après l’autre; ils forment une
chaîne” [the acoustic signs have only the time line; their elements
appear one after the other; they form a chain] (103). Given the views
above, we can say that speech develops gradually being built up with
signs, that is, units endowed with a sense. A speaker articulates
his/her thought using successively emerging signs. A required condition
to understand it is mapping sound forms to a content more or less
identical for both speaker and addressee. A listening addressee hears
the speech as an increasing by every sign speech segment subjected to
understanding. The addressee gets a partial understanding of speech as
it gradually develops in attempting preliminary guessing of the meaning
of the sentence while anticipating the resulting construction completion
for full understanding the sentence.
In constructional semantics as Ada Rohde (2001) says, “In general,
constructions are seen as form-meaning pairings that link a specific
syntactic structure to abstract meaning components. Constructions are
thus not different in kind from lexical items, they are simply more
abstract representations. … One of the defining properties of
constructions is that their meaning is not entirely predictable from
their component parts. Strict compositionality is thus excluded by
definition” (37). Thomas Hoffmann and Graeme Trousdale (2013) notice
that “… the analysis of syntactic structures as form-meaning pairings
was commonplace in traditional grammars. … Consequently, he [Paul
Kay] makes a distinction between constructions as fully productive
processes that are part of a speaker’s grammar, on the one hand (such as
the All-cleft construction), and semiregular processes such as the ’A as
NP’-pattern (as exemplified by dumb as an ox, dead as a doornail, or
green as grass), on the other, which he sees as mere patterns of coining
(and which are part of metagrammar, i.e. ”a compendium of useful
statements about the grammar”).” In this regard, let us note for now
that when meanings enter a construction in speech, not only meanings but
also semantic information is involved (see also R. Jackendoff above).
The relation of information and meaning is operated from quite different
angles. Loet Leydesdorff (2016) mentions that “As is well known,
Shannon [Shannon, C. E. 1948. A Mathematical Theory of Communication.
Bell System Technical Journal, 27, 379-423 and 623-656] first focused
on information that was not (yet) meaningful: ʻFrequently the messages
have meaning; that is they refer to or are correlated to some system
with certain physical or conceptual entitiesʼ (3). … one can
distinguish between “meaningful information” — potentially reducing
uncertainty — and Shannon-type information that is by definition equal
to uncertainty [Hayles, N. K. 1990. Chaos Bound; Orderly Disorder in
Contemporary Literature and Science Ithaca, etc.: Cornell University, p.
59]. … Meaning is provided to the information from the
perspective of hindsight (of the “later event” — that is, a system
of reference) (4). On the one hand, Luhmann [Luhmann, N. 1995. Social
Systems. Stanford, CA: Stanford University Press, p. 67] defined
information as a selective operation and stated that “all information
has meaning. (15)” “As will be shown in the following definitions of
information concerning Wiener and Shannon,” writes Kane X. Faucher
(2013), “none of these have anything to do with knowledge claims and or
semantic meaning; they are largely mathematical concepts. …
information is made material when incarnated in artefacts,
objects, and entities. In this way, information is what “haunts”
matter, while depending on it (8).” According to John Mingers (1995)
“… meaning is created from the information carried by
signs . The consequences are that information is objective, but
ultimately inaccessible to humans, who exclusively inhabit a world of
meaning. Meaning is essentially intersubjective – that is, it is
based on a shared consensual understanding (3). This explanation reveals
the relationship between information and meaning (import) – objectiveinformation is converted into (inter)-subjective meaningthrough a process of digitalization (11). Information is an objective,
although abstract, feature of the world in the same way as are physical
objects and their properties (12)”. Our own understanding is different
from the cited ones. The choice of our notion of linguistic information
is based upon the phenomenon of disconnecting meaning from the word in
speech (see further next paragraph): The meaning detached from the word
ceases to be meaning but remains information. It is also necessary to
draw a distinction between meaning as a means of identifying (the
denoting property is a characteristic not of a word as pure form but of
its meaning – it’s meaning that enables naming), which we can call
semantic information, and meaning as a subject of study. Semantic
information turns to external reality and is connected with our
knowledge of the world we live in, whereas in analyzing meaning as
subject of study we must reveal its own characteristics and not what it
denotes.
Finally, we would like to notice some important distinctions of the
proposed approach from the available linguistics studies. This research
is not part of traditional grammar, but it is also distinct from modern
theoretical grammar in object of study. In contrast to them, the
proposed approach deals with nonsemantic entities. But it can’t be
defined as a formal system either because, although the nonsemantic
entities don’t refer to the meaning of the words11See Rudolf
Carnap’s (1937: 1) definition of a formal theory:
A theory, a rule, a definition, or the like is to be calledformal systems when no reference is made in it either to the
meaning of the symbols (for example, the words) or to the sense of the
expressions (e.g., the sentences), but simply and solely to the kinds
and order of the symbols from which the expressions are constructed.
, they are embedded in linguistic information. In
Cognitive Linguistics, as Fauconnier (2014: 2) says, “Cognitive
capacities that play a fundamental role in the organization of language
are not specific to language”. Regardless of the Cognitive Linguistics
findings, the studied level formed with operational structures (see
below) is specific to speech process. In Cognitive Linguistics, as
Geeraerts and Cuyckens (2007: 5) say, “Language is seen as a repository
of world knowledge, a structured collection of meaningful categories
that help us deal with new experiences and store information about old
ones”. What interests us here, is just how the available meanings
interact in speech through parametric characteristics and not how a
meaning correlates with other meanings or knowledge.