How to Cite:

Petrova, I., Sezonov, V., Perlin, S., Sezonova, O., & Piddybna, A. (2021). Linguistic
document research technologies (forensic and procedural aspects). Linguistics and Culture
Review, 5(S4), 1464-1482. https://doi.org/10.21744/lingcure.v5nS4.1767

Linguistic Document Research Technologies

(Forensic and Procedural Aspects)

Iryna Petrova
National Scientific Center “Hon. Prof. M. S. Bokarius Forensic Science Institute”,
Kharkiv, Ukraine

Viktor Sezonov
Kharkiv Scientific Research Forensic Center of the Ministry of Internal Affairs of
Ukraine, Kharkiv, Ukraine

Stanislav Perlin
Kharkiv Scientific Research Forensic Center of the Ministry of Internal Affairs of
Ukraine, Kharkiv, Ukraine

Olga Sezonova
Kharkiv National University of Internal Affairs, Kharkiv, Ukraine

Alina Piddybna
Kharkiv National University of Internal Affairs, Kharkiv, Ukraine

Abstract---Linguistic examination is a modern type of forensic

examination, which is directly related to the study of the products of
speech activity in order to establish facts (circumstances) of
evidentiary significance in a particular (criminal, civil and
administrative) case. Although linguistic knowledge is often useful,
author identification has not yet reached a significant level of
reliability. The main purpose of the study is to investigate the current
state of legal regulation of document linguistic examination in Ukraine
and to disclose the criminal and procedural aspects of linguistic
technologies of document examination. The research process involved
using a combination of general scientific, philosophical and
specialised scientific methods, which are in line with the aims and
objectives set by the authors. The philosophical basis of the
methodological paradigm of the study is the dialectic as a
philosophical learning method. The author's conclusion provided a
conceptual understanding of the implementation of document
linguistic examination in Ukraine. The review of implementation and
evolution of legal regulation in the outlined area demonstrates that
Ukraine in directions of legal regulations reform attempts to be

Linguistics and Culture Review © 2021.

Corresponding author: Petrova, I.; Email: petrova7063-1@murdoch.in
Manuscript submitted: 09 July 2021, Manuscript revised: 27 Oct 2021, Accepted for publication: 18 Nov 2021
1464
 1465

oriented towards implementation of progressive international

approaches to forensic linguistic examination, paying attention to the
urgent need to develop methodological approaches to examination of
physical products containing destructive information due to the
growing threats of virtual space. The relevant legal mechanisms
suggested by the scholars can be generalized, adapted and
implemented into the Ukrainian legislation, which regulates the
procedure of conducting linguistic examination of documents. The
study suggests the gradual introduction of positive foreign practices of
adapting innovative technologies of linguistic research to the domestic
realities of today.

Keywords---document examination, expert technology, forensic

evaluation, linguistic examination, procedural.

Introduction

Forensic linguistic examination has become one of the practices that is developing
rapidly within the judicial process in the third millennium, and this trend is due
to the dramatic increase in multimodal communication among individuals and
organisations (Seyari & Bagheri, 2019). Forensic linguistics is an interdisciplinary
field of applied linguistics that mainly analyses language in two directions
(Ahmed, 2021). Firstly, from the perspective of offence, it may be useful in the
investigation of crimes. An integral part of forensic linguistics in this field is the
identification of the author of written or spoken texts. Forensic linguistics can
thus be called the interface between language and law (Ahmed & Arcelus-
Ulibarrena, 2021). The second area of focus is the research and use of language
for judicial procedures. Forensic linguistics is a subdivision of linguistics which is
particularly closely related to professional and institutional interaction in legal
contexts, it is the study of language in a legal framework, with texts, both oral
and written, forming the basis for the study, analysis and measurement of
language (Coulthard et al., 2016). Linguistic expertise, as a field of knowledge that
studies the understanding of language use in legal proceedings and develops
guidelines for the provision of linguistic expert opinions (Umiyati, 2020),
complements legal analysis by applying rigorous, scientifically accepted principles
of language analysis to legal evidence such as emails, text messages, contracts,
letters, confessions and recorded speech (Leonard et al., 2017). Nowadays, the
need to simplify the processes of mechanical processing of arrays of information,
including its systematisation, search automation, translation into other
languages, annotation, has stimulated the emergence of computer linguistics; the
development of optimal models of mass management and has confirmed the need
to recognise ways of influencing the recipients' minds, which, in turn, leads to the
justification of psycholinguistic approaches as an explanatory tool of science.

Linguists now apply their knowledge in areas such as: identification of the author
of a written document or the speaker of an audio document, police interrogation
practices, contract disputes, legal discourse, defamation, trademark infringement,
copyright disputes, discrimination, commercial warning messages and various
types of criminal charges such as bribery, extortion, money laundering, threats
1466

and fraud. Practically all such cases involve written or oral evidence, which makes
their linguistic analysis extremely relevant. In contract disputes, the meanings of
individual words and phrases (as well as syntactic relations) can form disputes. In
plagiarism cases, which are a subset of copyright analysis, the question arises
whether the defendant has taken text or content from an author or company
document (such as a novel, a court opinion, a screenplay or a patent applications)
to another document without proper citation (Pfeffferli, 1983; Biedermann et al.,
2011). In copyright cases, linguistic issues may include not only simple borrowing
of words but also copied discourse structure, such as a sequence of themes. In
the relevant branch of the law, trademark infringement cases regularly involve
linguistic similarities between a junior and a senior trademark (for example,
phonological analysis can demonstrate that they sound similar, and semantic and
pragmatic analysis can ascertain similar meanings). Even in cases of liability for
copying a certain product, linguists will help to confirm important information, for
instance by revealing that the product contained insufficient, incomprehensible
information. Other types of cases in linguistic analysis can become key ones –
these are cases of discrimination and defamation, where the defendant's speech
can be subject to revision, for example by its meaning in context. In today's
realities, "hate speech" becomes a type of malicious online content that directly
attacks or promotes hatred towards a group or individual member based on their
actual or assumed aspects of identity, such as ethnicity, religion and sexual
orientation. With hate speech on the Internet, there is a growing interest in its
automatic detection as a natural language processing task (Yin & Zubiaga, 2021).

The information obtained as a result of linguistic examination of documents must

be used in the course of procedural evidence, so it must be materialised in a
procedure and form defined by law – the means of evidence. Therefore, the issue
of studying different sources of evidentiary information, especially new ones
related to electronic means of evidence, is always relevant (Kalamaiko, 2016).
Courts recognise the legitimacy of the field of forensic linguistics and entrust
experts to give opinions and, in some cases, evidence. This position applies to a
wide range of cases and situations. Nevertheless, although forensic linguistics has
been widely introduced into judicial practice, it is still an underused tool; at the
same time it can be applied to almost any case in which speech may be
considered as evidence, and this certainly covers many more cases than those
where it is actively tested today (Leonard et al., 2017).

The social demand for conducting forensic linguistic examinations as required by

the current Ukrainian procedural legislation objectively required professional
expert and linguistic support for court proceedings. Since the number of
information and documentary disputes has increased, involving not only citizens,
legal persons, but also the media and Internet publications, an objective need has
emerged to transfer this examination to the level of a well-tested technological
process, which should be carried out using unique, scientifically validated
methods and reproduced linguistic technologies (Yaroshchuk et al., 2020). Expert
technology in this sense is an information model that includes the task of expert
investigation, methods and techniques for solving these tasks in order to establish
the factual data in the process of forensic investigation (Korma, 2017). However,
specialists still lack a unified, systematically organised and empirically grounded
base of diagnostic and identification features necessary to make fundamental
 1467

knowledge in linguistics the basis for the development of tested technologies for
solving practical forensic tasks.
The conclusions of an expert are recognised as evidence in the course of
investigating a crime. They are independent of the procedural status of the person
who initiated the examination, have no pre-determined force or advantages over
other evidence, and are subject to verification, procedural and forensic
assessment by the prosecution, the defence and the court (Shcherbakovskyi,
2013). Thus, 82 percent of the interviewed investigators and 74 percent of the
judges consider expert opinion to be a stronger source of evidence than other
evidence (Shcherbakovsky, 2015). The investigator, prosecutor and court may
recognise the findings of the expert examination as admissible evidence if the
requirements of the criminal procedure law and regulations for the appointment
of the expert examination, the research methodology of the expert examination
have been complied with and the findings contain reliable evidentiary information.
On the basis of the aforementioned, it is highly relevant to consider key issues
concerning the technology of linguistic research of documents. In order to achieve
the objective in view, it is necessary to define the following tasks: 1) to consider
the main points of the state of modern legal regulation regarding the
implementation of speech linguistic examination in Ukraine (forensic and
procedural aspects); 2) to consider a number of lawyers' recommendations
regarding the implementation of technologies for linguistic examination of
documents with the aim of their possible implementation in the relevant
procedures in Ukraine.

Materials and Methods

The methodological basis of the study was a set of subject-based principles,

approaches and methods of knowledge. A combination of general scientific,
philosophical and special methods of science was used in the research process,
which corresponds to the purposes and objectives of the study. Thus, general
scientific methods such as analysis and synthesis, induction and deduction,
comparison, abstraction, etc. were used in the process of writing the paper.

Conducting a gradual historical and legal analysis of the selected topic is

impossible without taking into account the transformations that have occurred
not only with the object of study, but also with all the processes and phenomena
related to it. The mentioned above, primarily, enables to identify and take into
account all factors and conditions that have determined the evolution of
technologies for linguistic research of texts in Ukraine, especially in the context of
forensic and procedural aspects, and so the historical and legal method is used to
determine the stages, analogues and determinants of international cooperation in
the field of improving the legal regulation of the relevant linguistic studies; the
problem-chronological method enabled the research text to be structured, and the
empirical analysis facilitated comparison of historical facts and competent expert
characteristics.

The leading method of scientific research was the dialectical method, based on
two principles: 1) determinism (the principle of conditionality and
interrelationship between different phenomena) 2) historicism (any phenomenon
is not static, but develops in a temporal dimension). The method chosen made it
1468

possible to focus scientific research on transformations in the technology of

linguistic document research through the prism of societal development and
qualitative changes in the contemporary realities of legal consciousness and law
enforcement. This method also enabled the identification of the state, directions
and prospects of development of scientific research and legislative developments
in the field of legal regulation of the conduct and significance of document
forensic linguistic examination.

The methodology of the study is also based on a systematic approach, due to the
specific nature of the paper and related to the use of general scientific and
specifically scientific methods. The historical method is used to investigate the
formation and development of complex forensic linguistic studies, the stages of
their formation and the research technology. The formal-logical method used
when working on the study made it possible to determine the logic of forensic
expert linguistic research, its specificity, the availability of special knowledge
among experts and the specifics of implementing expert conclusions in this field.
Statistical and comparative legal methods were used to identify areas of
improvement in the legal framework and the practice of applying the law
regulations governing these issues by forensic experts and the persons
(authorities) who order linguistic research of documents.

The methods of grammatical examination, interpretation and comparison of

individual categories and legal provisions were also applied, helping to identify
gaps and other shortcomings in the current Ukrainian legislation and providing
suggestions for its improvement. The method of legal analysis was used when
studying the relevant provisions of forensic science, criminal and civil procedural
legislation, departmental regulations on the issues of expert support of justice
and preliminary investigation; the forecasting method – when developing
suggestions for improving the current legislation in the field of linguistic research
of documents, the practice of its application.

Induction enabled general conclusions to be drawn on trends in the development

of document linguistic research technology by studying specific statistical data.
The deductive method was applied when dealing with legal provisions that directly
regulate the admission of the results of document linguistic research as proper
and admissible evidence in various proceedings. The specific sociological method
used in the study of law enforcement materials made it possible to assess the real
implementation of legal regulations and basic technological aspects of conducting
research with texts in the life of society, the efficiency of legal regulation in the
studied area, the impact of the results of linguistic research on law enforcement
practice and the state of law and order, on the qualitative characteristics of legal
culture and legal consciousness. As part of the analysis of legislative recognition
of the results of linguistic research on documents, the following ways of
interpreting the law were taken into account: logical, systematic, grammatical,
historical and functional.

Nowadays, with globalisation and increasing legal ties between different states,
the importance of the comparative method is constantly growing, as it holds great
methodological potential. For this reason, the study used the special legal method
of knowledge – the comparative law method (comparative method). In particular,
 1469

macro comparisons were used in the context of the significance of testing the
latest foreign technologies of linguistic research of documents in Ukraine.

Results

In Ukraine, the commissioning and conducting of forensic examinations and

expert studies (Law of Ukraine No. 4038-XII, 1994) is regulated by the relevant
departmental instructions under which linguistic examination of speech is a
component of forensic expertise and is carried out by forensic experts. This
forensic expertise is performed by state specialised institutions (Law of Ukraine
No. 4038-XII, 1994), which include research institutions of forensic expertise of
the Ministry of Justice of Ukraine; research institutions of forensic expertise,
forensic medical and forensic psychiatric institutions of the Ministry of Health of
Ukraine; expert services of the Ministry of Internal Affairs of Ukraine, Ministry of
Defence of Ukraine, Security Service of Ukraine and State Border Guard Service of
Ukraine. According to the Scientific and Methodological Guidelines on the issues
of preparation and commissioning of forensic examinations and expert studies
(Order of the Ministry of Justice of Ukraine No. 53/5, 1998), the linguistic
examination of speech includes linguistic examination of written and oral speech,
with the objects of examination being the products of human speech activity
reflected in written and oral form, fixed in the (audio, video) recordings. It is worth
noting that since 2017 Ukraine has introduced the legal institution of electronic
evidence in the Economic Procedural Code of Ukraine, the Civil Procedural Code
of Ukraine (2012) and the Code of Administrative Procedure of Ukraine (2017).

The modern division of linguistic examination according to the complex "object –

subject – expert technology" framework is as follows: naming expertise
(examination of names); linguistic examination of regulatory legal acts and
documents; linguistic examination of advertising texts; extremist materials;
defamatory materials (Chang et al., 2008; Wen, 2010). Linguistic examination is
most often commissioned in cases of bribery, extortion, threats, drug trafficking,
etc. (Nikishin, 2018). International linguists suggest the following tasks to be
applied in the linguistic examination of speech (The Arts and Humanities
Research…, 2010): Identification of the author, includes determining whether a
particular person has said or written something, and which is based on an
analysis of their idiolect or particular patterns of language use (vocabulary, word
combinations, pronunciation, spelling, grammar); forensic stylistics, which
subjects written or spoken materials (or both) to scientific analysis in order to
determine and assess content, meaning, identify the speaker or identify the
author when plagiarism is detected; discourse analysis concerns analysis of
written, spoken or sign language as well as any meaningful semiotic event;
linguistic dialectology – the methodological study of dialects based on
anthropological information; forensic phonetics deals with producing an accurate
transcription of what is said, which can reveal information on the social and
regional origin of the speaker; forensic transcription in the form of written
documents as well as video and audio recordings; within-author variations – ways
in which texts by the same author differ from one another (Madea & Preuß, 2009;
Bataille et al., 1999).
1470

The object of expert linguistic research are written materials of various types:
newspaper and magazine publications; speech scripts; oral speech scripts of
participants in legally significant communication situations; books; legally
binding documents (contracts, receipts), case materials for which linguistic
examination is conducted (witness statements, applications, minutes, etc.),
Internet communication texts (forums, blogs, social networks). The subject of the
examination is the facts and circumstances of the case established on the basis of
an examination of the regularities in the existence and use of a natural or
artificial language by its speakers (Gow Jr & Im, 2004; Waxman et al., 1997).

The tasks of linguistic examination include: scientific interpretation and

clarification of the meaning and origin of a word, word combination, phraseology
or other linguistic unit; interpretation of the primary and secondary (connotative)
meaning of a linguistic unit or a linguistic unit actualised in context; investigation
of a text (fragment) to identify its semantic orientation, modality of sentences,
expressiveness and emotionality of linguistic units, their formal and grammatical
characteristics and semantics, the specifics of the stylistic means and techniques
used. The tasks mentioned are specified according to the field of application of
special linguistic knowledge in different case categories. Unlike the examination of
audio speech, which identifies the acoustic characteristics of language utterances,
including identifying the author of a speech message, the written text is analysed
according to a range of features: grammatical, semantic, pragmatic, which
involves the use of traditional methods and approaches grounded in linguistic
science. Linguistic examination of the text establishes the relationship between
the person, the level of language, communicative writing competence,
characteristics, purpose and effects of language; it makes full use of all the
achievements of basic science: the concept of language identity, typology of texts,
literary language and spontaneous speech, the correlation of written and spoken
forms of expression, norms and development in the process of communication,
etc. The result of a linguistic examination is the opinion of an expert, the main
part of which is structured as a response to court enquiries or an information
dispute party and contains conclusions on the disputed expressions of a
conflictogenic text or on the text as a whole (Bogoslavska, 2020).

The Ukrainian procedural legislation regulates the activity of experts by types of

examination, which include primary, additional, repeated, commission and
complex examinations. The relevant expert investigations are carried out with the
help of special knowledge and the use of methods of forensic science and forensic
examination (Ullman, 2004). The basis for conducting an examination should be:
a procedural document appointing the examination, or an agreement with an
expert or an expert institution, with mandatory references to the relevant articles
of law; a list of issues to be solved; and the objects to be examined (Ullman &
Pierpont, 2005). The results of expert examinations and expert studies shall be
provided in a written document – expert opinion, and the period of expert studies
shall not exceed 90 calendar days. Linguistic experts should not draw
conclusions about legal issues. The competence of the expert linguist does not
include the legal assessment of the texts examined, i.e. the expert does not draw
conclusions about the presence of insults, slander, humiliation of the honour and
dignity of a person (group of persons), about obscene word forms, etc., and does
not classify materials as extremist, does not establish the extremism signs.
 1471

Regardless of how strongly the linguist is convinced that the defendant is

innocent, he or she should limit his or her opinion to stating the degree of
probability. Moreover, forensic linguists must always remain impartial, as they
implement the law in the role of experts and cannot, under any circumstances,
take the side of either the defence or the prosecution. In this case, the expert
needs to find out what form the information is in and provide a statement or
value judgement.

The organisation of conducting forensic examinations and registration of their

results shall be carried out in the manner prescribed by the Criminal Procedural
Code, the Civil Procedural Code, the Economic Procedural Code of Ukraine, the
Code of Ukraine on Administrative Offences, the Code of Administrative Court
Procedure of Ukraine, the Laws of Ukraine "On Forensic Examination" and "On
Enforcement Proceedings". Thus, for instance, with regard to criminal proceedings
in accordance with the requirements of Article 94 of the Code of Criminal
Procedure of Ukraine (2012), the conclusions of the expert as a source of evidence
for the crime under investigation are subject to procedural assessment from the
viewpoint of relevance, admissibility and credibility (Peniro & Cyntas, 2019).
According to part 2 of Article 84 of the Code of Criminal Procedure of Ukraine
(2012), the procedural sources of evidence are testimonies, material evidence,
documents, and expert opinions (Gabidullina et al., 2021). The documents
submitted for linguistic expertise serve as physical evidence in accordance with
Article 98 of the Code of Criminal Procedure of Ukraine (2012). In order to clarify
the validity of the facts stated in the expert opinion, it is necessary to analyse the
course and results of the expert examination. The investigator, prosecutor and
court may recognise the findings of an expert examination as reliable evidence if
the expert has conducted a full examination of the object within the limits of his
or her task and has provided reasonable and objective written conclusions the
evidential value of which directly or indirectly confirms the facts, circumstances
relevant to the investigation of the crime.

Linguistic examination of written speech in Ukraine is divided into two types –

authorship and semantic-textual investigations. Linguistic examination of spoken
language consists of identification and diagnostic studies of a person's oral
speech and semantic studies of spoken language (Suryasa, 2019). In accordance
with the tasks set, the linguistic expert uses standard questions, algorithms and
methods of examination. With regard to authorship examination, linguists study,
for example, grammatical structures in which changes in punctuation schemes
between texts can indicate different authors; semantics, which investigates how
meaning is formed, for instance when understanding written text;
sociolinguistics, which analyses differences in the use of language among different
social groups. The procedure for analysing the linguistic features of a text, as
adopted in linguistic expertology, includes the following aspects of studying its
features: lexical-phraseological characteristic features, grammatical and stylistic
features. When features are present at any one or more levels of analysis, the
question of their relevance is decided: the mass, the persistence that
distinguishes the abilities. A text of at least 100 words in the form of free,
conditionally free and experimental samples of the written speech of the person to
be identified must be provided for the purposes of the authorship examination.
Attributional methodology is often used, combining qualitative and quantitative
1472

approaches to text attribution. Qualitative analysis is a study of an author's

individual style as a holistic construct, which represents the linguistic
competencies of the author of a text and the level of proficiency in them
(Yunusova, 2021). With this type of analysis it is possible to create a linguistic
model of an author's individual style. Quantitative analysis is the extraction of
statistical data from a text array. This data helps to make the attribution
examination more complete, comprehensive and objective (Khomenko, 2019).

It is worth noting that in today's realities there is an increase in the number of

crimes involving digital technologies. These online offences are facilitated due to
the anonymity and reach allowed on social media platforms. Proper identification
of individuals who post threatening, defamatory or false messages online is of
paramount importance to investigators as it can help protect those who are
attacked. The authorship establishment is based on the correct grouping of texts
created by one author by highlighting textual features characteristic of that
author. These characteristics are usually related to grammatical structure and
are deeply rooted in each person's individual authorial style. Establishing
authorship in social media is challenging as there is no "text imprint" or clear
pattern of language use. Examples of text that are often analysed by forensic
linguists include blackmail letters, confessions, wills, suicide letters and
plagiarism, online texts such as online chat logs of sexually explicit content
between middle-aged men and underage girls, etc. Author profiling is used when
researching social bots, the most common of which are Twitter bots. Social bots
are considered a threat, given their commercial, political and ideological influence.
The bots designed to look like human accounts can mostly be identified by the
information in their profiles, such as username, profile picture and posting time.
However, the task of diagnosing bots based solely on textual data is a much more
complex one, requiring the methods of author profiling. This usually involves
classification tasks that are based on semantic and syntactic characteristics.

The subject of linguistic examination is also the determination of the text's

semantic content and the identification of criminally relevant information in the
text (Koutchadé et al., 2018). The linguistic investigations most in demand today
are those in cases of verbal imagery, protection of honour, dignity and business
reputation as well as in cases of extremist crime (Order of the Central Department
of the Security Service of Ukraine No 215, 2020). Extremist crime cases may
include the presence or absence in the text of a publication or document of a
threat to commit a particular crime (murder, terrorist act), calls to overthrow the
constitutional order, to seize state power, to commit deliberate acts with the aim
of changing the territory or state border, statements aimed at inciting national,
racial or religious hatred and enmity, expressions that can be classified as
propaganda for war, fascism, etc.

The materials are provided to the expert linguist for examination in the form of
printed texts, oral speech materials on digital media (CDs, flash cards, computer
hard drives). If the text under investigation is printed in a periodical as well as in
a book, a complete copy of the publication or a good quality copy thereof shall be
submitted for examination. In the case of a text document from the Internet, a
complete digital copy of the page(s) containing the text for examination, including
illustrations, or a printout of the page(s) in question, indicating the moment when
 1473

the seizure was made, should be submitted for examination (Sokolova-Vysochyna,

2018). In addition to verbal data, the texts under examination may include non-
verbal components (graphics, symbols, drawings, photographs, etc.), which also
serve as objects of linguistic examination due to the fact that they contain
additional important information. The expert also has to be provided with a
protocol of inspection and listening to the audio recordings of the conversations
under investigation with their typed text, drawn up in accordance with the
procedural requirements.

The texts of oral speech (speeches at rallies, conferences, interviews) can be

provided for linguistic investigation. The amount of language material should be
approximately 5-10 minutes of the test person's speech. When it comes to
linguistic examination of spoken language, speaker identification is the most
common task of the relevant investigation. This technology involves comparing
language models to assess the speaker's identifiability. Comparisons are made by
a comprehensive examination of the phonetic, linguistic and acoustic components
of speech and can be supplemented by automated procedures using speaker
recognition technology. A comprehensive examination of an oral audio recording
is carried out jointly by an expert linguist and an expert engineer in the field of
video and sound recording. The examination of oral statements involves analysing
problem parts of the recordings to determine what was said. This can be an
argument, for example, through the noisy environment, overlapping language,
accents, dialects. Improvements can be made to help with this type of work,
which include the use of digital filters and dynamic processors in recordings to
reduce background noise and improve speech intelligibility. Profiling technology
involves analysing the recordings of language unknowns to obtain information on
the regional and social origin of the speaker. In this case, background noise can
also provide information on where and when the recording was made. Forensic
linguists can analyse interrogations recorded by the police, which can be used to
decipher the facts about a person's guilty plea – whether or not it was consciously
admitted. It may also be established that the person "only underwent
interrogation" or only "understood the conversation" during the investigation or
interrogation. Recorded interrogations are presented as important evidence in
court. Thus, the technique of dialogic interview analysis is used by forensic
linguists who can prove someone's guilt or innocence. This analysis can also
identify potential inconsistencies in the interview process, making the recordings
inadmissible evidence in court. The defence can thus demonstrate that the
recorded speech does not necessarily indicate the guilt of the defendant.

Increasingly, forensic linguistic examination uses a combination of software,

experience and statistical approaches in its analysis. Computer scientists have
developed technologies to automate linguistic and phonetic analysis. These
approaches do not require an expert to implement them, but do require expert
interpretation (Forensic Language Analysis, 2015). Computational technology
extracts biometric information (based on the physiology of an individual's vocal
tract) from speech samples. These samples can be compared with others to
perform an automatic comparison (sometimes known as recognition) or to check
whether the same person is speaking in several samples (verification).
1474

Technological advances in recent decades have opened up new possibilities for

forensic linguistic analysis: new forms of online interaction have required more
advanced forms of computer-mediated discourse analysis as well as synchronous
and direct forms of communication, such as those provided on online platforms,
which have enabled users to communicate with almost anyone, anywhere, at any
time, from any mobile device using online communication (Sousa-Silva, 2019). In
this case, corpus linguistics technology is applied to the analysis of meaning
(written and spoken): software processes hundreds of documents such as
extremist texts on the Internet and identifies key words, phrases and themes that
can be used for intelligence gathering and investigative purposes.

Knowing the author of the information can help better determine its reliability,
which is critical in light of the increasing amount of online misinformation spread
by so-called trolls, bots and other online agitators. A great deal of work on
implementing machine learning algorithms by computational and statistical
methods to determine authorship of textual works based on writing style: word
choice, use of punctuation marks, idiosyncratic grammatical errors, and in the
latest digital texts – the use of emoticons – is carried out by, for example, PAN,
PolEval – active international communities in which computational approaches to
author identification are developed and evaluated (Ordoñez et al., 2020).

Based on the aforesaid, it can be concluded that at the present stage of

development of Ukraine the activity of forensic experts in conducting linguistic
examination of speech is regulated at the legislative level. The relevant expert
actions are carried out with the help of special knowledge and the use of methods
of forensic science and forensic examination. In procedural terms, the
conclusions of an expert investigation are considered reliable evidence, if the
expert has conducted a full examination of the object within the scope of his or
her task, and has provided sound and objective written conclusions, the
evidential value of which directly or indirectly establishes the facts, circumstances
relevant to the investigation of the crime. The linguistic investigations most in
demand today are those in cases of verbal imagery, protection of honour, dignity
and business reputation as well as in cases of extremist crime. Increasingly,
forensic linguistic examination uses a combination of software, experience and
statistical approaches in its analysis.

Discussion

Forensic linguistic analysis can be of value in almost any case where oral or
written speech utterances may be considered as evidence (Leonard et al., 2017).
Recently, researchers have emphasised the important role of linguists involved in
forensic linguistics (Houtman & Suryati, 2018), paying special attention to the
interdisciplinary nature of the investigations and highlighting the importance of
pragmatics as the linguistic basis of professional activity. Moreover, practitioners
emphasise the importance of collecting and providing accurate data in terms of
applied characteristics (Meluzzi et al., 2020). Conclusions based on the performed
linguistic analysis of documents acquire the status of reliable evidence through
further research of disputed meanings, the application of authorship analysis
techniques in response to new needs (e.g. cybercrime investigations) and attempts
 1475

to develop new theories, such as synthesis of author speech (Grant & MacLeod,
2018).

An important aspect of forensic linguistics is the identification of the author of a

particular written text. Certain problems should be highlighted in this aspect.
Thus, documents available for forensic analysis are sometimes very short
extracts. These can be ransom notes or letters containing blackmail materials.
Consequently, these documents can hardly be used as reliable identification of
the author (Ahmed, 2021). Another problem is that it remains to be identified and
determined which linguistic features can be reliable indicators for determining
authorship. The credibility of linguistic characteristics is also a big issue in the
field of forensic linguistics. There is an acute need to identify reliable attribution
methods for the application in short texts. However, several important issues
remain open regarding "author's indication", in which the most important issue is
the required length of the text. Various investigations and reports have shown
fruitful results with short criminal texts that are less than 1000 words in length,
but forensic linguists have not yet been able to determine the minimum text
length that can be recognised as reliable to establish the author (Najam-Us-
Sahar, 2020).

One of the controversies in the application of forensic linguistics is the divergence

identified among forensic linguists. This divergence relates to the accuracy of the
speaker's identification. There may be certain "vocal distortions" in the speaker's
voice sample provided for verification, and it can be very difficult for specialists to
deal with this. The probability of identification in such a sample is lower than in
conventional voice samples. This problem could be solved if the experts suggest to
the court to oblige the suspect to give a voice sample that repeats certain phrases
in a natural conversational voice (Ahmed & Arcelus-Ulibarrena, 2021). Voice
samples obtained through such specific recommendations are usually very
suitable for comparison purposes, but there is a universal standard for the
number of words required for identification. It can range from 10 to 20 words for
various situations (Machado et al., 2019). Using spectrographic voice
identification technology, it can be stated that the variability within the technical
means that records and reproduces oral speech differs from the variability
between different technical means. This means that differences in the same
expression of different speakers are noticeable, but scientific theory and data do
not consider this assumption to be confirming. There are certain points of view
based on the actual number of errors associated with professional judgements
about them. Moreover, "fragmented experimental results" also affect these
perceptions of the actual error rate. It should be noted that in this case "objective
data" is not given much significance, one could say that it is not representative of
the results when it comes to forensic examinations (University of York, 2021).

In terms of the international development of forensic linguistics, the following

challenges arise before attaining appropriate status: the integrated study of
forensic linguistics / language and law through different judicial systems and
geographical boundaries; the development of repeatable analysis techniques for
use in expert conclusions to ensure the internal and external validity of the
investigation; extensive detailing of codes of good practice and conduct;
cooperation of the International Association of Forensic Linguistics (IAFL) with
1476

other forensic science associations and societies (Ahmed & Arcelus-Ulibarrena,

2021). It will also be important for linguists in the era of international courts to
understand the practice of international law discourse and become familiar with
the customs of other countries' legal systems. It is likely that in the future an
increasing number of those seeking to enter the field of forensic linguistics will
have additional qualifications in areas such as law and will gain a better
understanding of scientific techniques, methods and presentations (Ariania et al.,
2014).

Forensic linguistics can be applied simultaneously in civil, administrative and

criminal cases to support legal conclusions; hence it is necessary to focus on each
aspect of the latest technologies that are changing traditional methods of
investigation and interrogation in pre-trial investigations. When oral or written
speech utterances can be regarded as evidence, every linguistic investigation can
be relevant (Syam, 2018). It should also be noted that there is an urgent need to
develop methodological approaches to the investigation of physical products
containing other destructive information due to such growing threats of virtual
space as: trolling, bullying, promotion of suicidal behaviour (including "death
groups" in social networks), propaganda of underground culture, the cult of
violence and cruelty (including prison culture), destructive propaganda using
"fan-fiction" content, etc. (Nikishin, 2018).

There is now an acute need to develop a methodology for the forensic authorship
examination of printed texts, taking into account the improvement of already
existing authorship examination techniques developed with handwritten texts
(Goloborodko, 2020). As a result of the fact that typed texts have features of
composition (automatic correction of errors, intervention by another individual
(performer) etc.), scholars insist that it is necessary to investigate and identify a
set of features that are stable for the author of a typed text, including graphic
ones (use of emoticons, other markings etc.). Assessing the validity and reliability
of corpus? procedures is very challenging, as data on the linguistic and phonetic
levels of the population are limited, making it difficult to establish the prevalence
of features. This means that conclusions cannot be expressed statistically or with
the same certainty as in other fields of forensic science. In linguistics, experts use
their knowledge and experience as well as computational methods in their
analysis; in phonetics, experts use their software to process the relevant sounds
and speech. However, since variability within the speaker makes each linguistic
expression unique, the expert cannot draw definite conclusions about, for
instance, the extremist orientation of texts. Computational procedures can be
used in some circumstances and give numerical conclusions that also express a
degree of certainty (Forensic Language Analysis, 2015).

Computational methods in linguistics have developed considerably in recent

decades. The increase in computing power of the computer, together with the
increasing attention of computer scientists to the processing of natural language,
has enabled more in-depth research into computational and computer-linguistic
analysis. Sophisticated computing systems and models have been developed that
allow large volumes of linguistic data to be analysed with little human
intervention, at a pace and degree of efficiency with which linguists can scarcely
compete (Sousa-Silva, 2019). However, according to the scholar, this does not
 1477

mean that linguists cannot be – or cannot act as – computational linguists, rather

the opposite; even if linguists do not have the advanced programming skills of
computer scientists, they have the knowledge necessary to assess the cost of
computing resources under certain circumstances and choose the most
appropriate computing tools for solving a particular linguistic problem. This is
particularly important in forensic contexts where linguists, in addition to
reporting their analysis, have to justify their findings scientifically and provide
transparency for court purposes.

The development of machine learning techniques and eventually artificial

intelligence (AI) raises new questions for forensic linguists. In addition to PAN
contests over the years, PolEval has organised a task aimed at identifying harmful
tweets in general and identifying the type of harm (cyberbullying or hate speech).
If, on the one hand, AI in particular will become increasingly competent at
producing human-like texts, then on the other hand (computational) forensic
linguists will face the need to develop, test and refine their methods and
techniques to solve the even greater problem of forensic linguistic examination
arising from the increasing complexity of computer systems (Sousa-Silva, 2019).
Since AI systems operate like black boxes, the results of their analysis cannot be
explained – and certainly not to the extent and with the level of transparency
required by the courts; however, they can play a key role in the context of an
investigation (Ionova, 2017). If machines are able to generate human-like text,
forensic linguists will have to be able to distinguish between human-generated
texts and machine-generated texts. Moreover, as the scholar notes, forensic
linguists may need help on cases of machine-generated text to establish whether
this text has some resemblance to the text production of whoever is controlling
the system, or vice versa, or whether the text was created by a machine to
resemble someone else's text.

In the case of evaluating electronic means of proof, it would be more efficient to

prepare for the process an electronic evidence bases on electronic media and in
printed form, provided that the relevant data allows for such actions (Kalamaiko,
2016). As a consequence, this situation will enable the court to establish the
affiliation of the evidence, and other persons involved in the process will also have
an opportunity to examine the case materials. The legal scholar identifies three
options for preserving electronic evidence and emphasises the need to avoid losing
important information for the case electronically and to avoid errors in procedural
implementation: out-of-court, contractual and judicial, and makes the following
suggestions: to increase the activities of the notarial system to certify the use of
electronic means of proof by amending the legal framework, which already has
relevant blanket rules; to introduce sanctions for spoiling evidence in the form of
monetary and court decisions on the existence of a specific fact in favour of the
other party into the legislation (Kalamaiko, 2016).

In order to solve the problem of assessing the admissibility and credibility of the
results of the expert examination, the investigator, prosecutor, defence counsel
and judge can use the following forensic recommendations: to find out whether
the expert has scientific, technical or other special knowledge necessary to give a
conclusion on the objects submitted for examination; to check whether the expert
has been warned of the relevant criminal responsibility; to check whether there
1478

are circumstances precluding the possibility of expert examination or whether the

expert is subject to recusal; to establish whether the expert understood the task
correctly and did not go beyond his or her competence; to find out whether the
expert informed the prosecution, defence or court about the impossibility to carry
out the examination due to lack of necessary knowledge or insufficient objects; to
find out whether the expert requested to provide additional materials and samples
for examination; to check whether the expert has received permission from the
initiator of the examination to fully or partially destroy the object of expert
examination or to change its properties; to check whether the expert has used
modern science and technology, means and tools, methods and techniques in
carrying out the examination; to find out whether he or she personally conducted
the examination of objects; to check whether the expert used special knowledge
when forming the conclusion; logically justified inferences based on the results of
the examination; to check whether the expert indicated in the conclusion the
information revealed during the examination, concerning which he or she received
no questions, but which were important for the investigation of the crime
(Shcherbakovskyi, 2013).

As a result of the discussion, it can be argued that the conclusions based on the
conducted linguistic analysis of documents acquire the status of reliable evidence
through further research into disputed meanings, the application of authorship
analysis techniques in response to new needs and an attempt to develop new
theories. It will be important for linguists in the era of international courts to
understand the practice of international law discourse and become familiar with
the customs of other countries' legal systems. It should also be noted that there is
an urgent need to develop methodological approaches to investigate physical
products containing other destructive information, due to the growing threats of
the virtual space. The investigator, prosecutor, defence counsel and judge can use
the forensic recommendations suggested by the scientists to solve the problem of
procedural verification of the results of the expert examination and the objectivity
of the conclusions.

Conclusions

Science is becoming increasingly important in relation to the law, and forensic

linguistics is one of the fields where investigations lead to advances that are
increasingly being used to uncover crimes. Forensic linguistics under many
circumstances provides powerful methods and techniques to substantiate
criminal charges or verdicts. Forensic linguistic expertise can also be used in civil
cases to support or appeal legal conclusions. At the present stage of Ukraine's
development, the activities of forensic experts in conducting linguistic
examination of speech are regulated at the legislative level. The relevant expert
investigations are carried out with the help of special knowledge and the use of
methods of forensic science and forensic examination.

In procedural terms, the conclusions of experts are considered reliable evidence, if

the expert has conducted a full examination of the object within the scope of his
or her task, and has provided sound and objective written conclusions, the
evidential value of which directly or indirectly establishes the facts, circumstances
relevant to the investigation of the crime. Moreover, the identification of
 1479

individuals who post threatening, defamatory or false messages online is of

paramount importance to investigators as it can help protect those who are
attacked.

When performing linguistic examination of written language in Ukraine,

authorship and semantic-textual studies are carried out. Linguistic examination
of spoken language involves identification and diagnostic studies of a person's
oral speech and semantic studies of spoken language. In accordance with the
tasks set, the linguistic expert uses standard techniques and algorithms for the
examination of documents. The linguistic investigations most in demand today
are those in cases of verbal imagery, protection of honour, dignity and business
reputation as well as in cases of extremist crime.

At the present stage in the implementation of linguistic examination, the following

technologies are mainly used: speaker comparison, examination of controversial
statements, profiling technology, technology for dialogic analysis of interviews,
technology for the automation of linguistic and phonetic analysis, corpus
linguistics technology. Increasingly, forensic linguistic examination uses a
combination of software, experience and statistical approaches in its analysis.
Furthermore, the study has shown that there is currently an acute need to
identify reliable attribution methods for use in short criminal texts.

Particular attention should be paid to the international development of forensic

linguistics, the practice of international law discourse and familiarity with the
customs of other countries' legal systems. It should be noted that there is an
urgent need to develop methodological approaches to investigate physical
products containing destructive information, due to the growing threats of the
virtual space. Therefore, there is now an acute need to develop a methodology for
the forensic authorship examination of printed texts, taking into account the
improvement of already existing authorship examination techniques developed
with handwritten texts.

According to scholars, in the case of evaluating electronic means of proof, it would

be more efficient to prepare for the process an electronic evidence base on
electronic media and in printed form, provided that the relevant data allows for
such actions. Legal scholars suggest the measures of providing electronic means
of proof to prevent the loss of case-relevant information in electronic form, as well
as for the purpose of its proper procedural design. The investigator, prosecutor,
defence counsel and judge can use the forensic recommendations suggested by
the scientists to solve the problem of procedural verification of the results of the
expert examination and the objectivity of the conclusions.

References

Ahmed, H. R. (2021). The Role of Forensic Linguistics in Crime Investigation: Uses

in Legal Proceedings. ANGLISTICUM. Journal of the Association-Institute for
English Language and American Studies, 10(2), 23-31.
Ahmed, H., & Arcelus-Ulibarrena, J. M. (2021). The role of forensic linguistics in
understanding legal documents in court proceedings: Controversies and future
1480

perspectives. Electronic Research Journal of Social Sciences and Humanities,

3(II).
Ariani, M. G., Sajedi, F., & Sajedi, M. (2014). Forensic linguistics: A brief overview
of the key elements. Procedia-Social and Behavioral Sciences, 158, 222-225.
https://doi.org/10.1016/j.sbspro.2014.12.078
Bataille, M., Crainic, K., Leterreux, M., Durigon, M., & de Mazancourt, P. (1999).
Multiplex amplification of mitochondrial DNA for human and species
identification in forensic evaluation. Forensic science international, 99(3), 165-
170. https://doi.org/10.1016/S0379-0738(98)00185-6
Biedermann, A., Taroni, F., Bozza, S., & Mazzella, W. D. (2011). Implementing
statistical learning methods through Bayesian networks (Part 2): Bayesian
evaluations for results of black toner analyses in forensic document
examination. Forensic science international, 204(1-3), 58-66.
https://doi.org/10.1016/j.forsciint.2010.05.001
Bogoslavska, M. O. (2020). Current issues of modern linguistic examination.
Theory and Practice of Forensic Science and Criminology: Materials of the III All-
Ukrainian Scientific-Practical Conference.
Chang, C. W., Wu, C. R., & Chen, H. C. (2008). Using expert technology to select
unstable slicing machine to control wafer slicing quality via fuzzy AHP. Expert
Systems with Applications, 34(3), 2210-2220.
https://doi.org/10.1016/j.eswa.2007.02.042
Coulthard, M., Johnson, A., & Wright, D. (2016). An introduction to forensic
linguistics: Language in evidence. Routledge.
Gabidullina, A., Sokolova, A., Kolesnichenko, E., Zharikova, M., & Shlapakov, O.
(2021). Metonymy in scientific linguistic discourse. Linguistics and Culture
Review, 5(S4), 71-83. https://doi.org/10.21744/lingcure.v5nS4.1556
Goloborodko, K. (2020). Linguistic examination of the text: Legal and linguistic
analysis. Ukrainian World in Scientific Paradigms: Collection of Scientific Works
of H.S. Skovoroda Kharkiv National Pedagogical University. Kharkiv: HIFT.
Gow Jr, D. W., & Im, A. M. (2004). A cross-linguistic examination of assimilation
context effects. Journal of Memory and Language, 51(2), 279-296.
https://doi.org/10.1016/j.jml.2004.05.004
Grant, T., & MacLeod, N. (2018). Resources and constraints in linguistic identity
performance–a theory of authorship. Language and Law/Linguagem e
Direito, 5(1), 80-96.
Houtman, H., & Suryati, S. (2018). The History of Forensic Linguistics as an
Assisting Tool in the Analysis of Legal Terms. Sriwijaya Law Review, 2(2), 215-
232.
Ionova, S. V. (2017). Aspects of the study of the written text as an object of
linguistic expertise. VolSU Bulletin. Series 2, Linguistics, 16(2), 28-38.
Kalamaiko, A. (2016). Electronic means of proof in civil proceedings.
Khomenko, A. Yu. (2019). Linguistic attributional study of short written texts:
qualitative and quantitative methods. Political Linguistics, 2(74).
Korma, V. D. (2017). Scientific foundations of criminology. Moscow: Prospekt.
Koutchadé, I. S., Datondji, A. C., & Salami, A. (2018). Exploring textual meaning
in Chukwuemeka Ike’s Sunset at Dawn: a systemic functional
approach. International Journal of Linguistics, Literature and Culture, 5(1), 1-11.
https://doi.org/10.21744/ijllc.v5n1.475
 1481

Leonard, R. A., Ford, J. E. R., & Christensen, T. K. (2017). Forensic linguistics:

Applying the science of linguistics to issues of the law. Hofstra Law Review,
45(3), Article number 11.
Machado, T. J., Vieira Filho, J., & de Oliveira, M. A. (2019). Forensic speaker
verification using ordinary least squares. Sensors, 19(20), 4385.
Madea, B., & Preuß, J. (2009). Medical malpractice as reflected by the forensic
evaluation of 4450 autopsies. Forensic science international, 190(1-3), 58-66.
https://doi.org/10.1016/j.forsciint.2009.05.013
Meluzzi, C., Cenceschi, S., & Trivillini, A. (2020). Data in forensic phonetics from
theory to practice. TEANGA, the Journal of the Irish Association for Applied
Linguistics, 27, 65-78.
Najam-Us-Sahar. (2020). A critical forensic linguistic analysis of media discourse
on court decisions in Pakistan: thesis for the degree of master of philosophy.
Islamabad: International Islamic University Islamabad.
Nikishin, V. D. (2018). The place of forensic examination of materials of a religious
nature of extremist-terrorist orientation in the classification of forensic
examinations. Journal of the OE Kutafin University, 7/2018, 187-198.
Ordoñez, J., Soto, R. R., & Chen, B. Y. (2020). Will longformers PAN out for
authorship verification. Working Notes of CLEF.
Peniro, R. ., & Cyntas, J. . (2019). Applied linguistics theory and
application. Linguistics and Culture Review, 3(1), 1-13.
https://doi.org/10.21744/lingcure.v3n1.7
Pfeffferli, P. W. (1983). Application of microspectrophotometry in document
examination. Forensic science international, 23(2-3), 129-136.
https://doi.org/10.1016/0379-0738(83)90142-1
Seyari, A., & Bagheri, M. S. (2019). Perspectives of Forensic Linguistics Research
in Iran. Journal of Language Horizons, 3(1), 187-205.
Shcherbakovsky, M. G. (2015). Conducting and using forensic examinations in
criminal proceedings. Kharkiv: V dele.
Shcherbakovskyi, M. H. (2013). Conditions for the formation of a reliable expert
report. Theory and practice of forensic science and criminalistics. Kharkiv:
Pravo.
Sokolova-Vysochyna, Ya. A. (2018). Appointment of linguistic examination of
speech in the investigation of crimes against the foundations of national
security of Ukraine. In: O. G. Ruvin (Ed.), Current issues of standardization of
forensic justice in Ukraine. Development prospects: Materials of the international
scientific-practical conference, dedicated 105th anniversary of forensic
examination in Ukraine and 95th anniversary of the birth of Academician M. Ya.
Segai. Kyiv: Kyiv Scientific Research Institute of Forensic Expertise of the
Ministry of Justice of Ukraine.
Sousa-Silva, R. (2019). Computational forensic linguistics: an overview of
computational applications in forensic contexts. Language and
Law/Linguagem e Direito, 5(2), 118-143.
Suryasa, W. (2019). Historical Religion Dynamics: Phenomenon in Bali
Island. Journal of Advanced Research in Dynamical and Control Systems, 11(6),
1679-1685.
Syam, S. K. M. A. (2018). Aspects of forensic linguistics in policing. Language in
India, 18, 100-111.
1482

Ullman, M. T. (2004). Contributions of memory circuits to language: The

declarative/procedural model. Cognition, 92(1-2), 231-270.
https://doi.org/10.1016/j.cognition.2003.10.008
Ullman, M. T., & Pierpont, E. I. (2005). Specific language impairment is not
specific to language: The procedural deficit hypothesis. Cortex, 41(3), 399-433.
https://doi.org/10.1016/S0010-9452(08)70276-4
Umiyati, M. (2020). A Literature Review of Forensic Linguistics. IJFL (International
Journal of Forensic Linguistic), 1(1), 23-29.
Waxman, S. R., Senghas, A., & Benveniste, S. (1997). A cross-linguistic
examination of the noun-category bias: Its existence and specificity in French-
and Spanish-speaking preschool-aged children. Cognitive Psychology, 32(3),
183-218. https://doi.org/10.1006/cogp.1997.0650
Wen, W. (2010). An intelligent traffic management expert system with RFID
technology. Expert Systems with Applications, 37(4), 3024-3035.
https://doi.org/10.1016/j.eswa.2009.09.030
Yaroshchuk, I. A., Zhukova, N. A., & Dolzhenko, N. I. (2020). Linguistic
examination. Belgorod: BelGU Publishing House, BelSU Research Institute.
Yin, W., & Zubiaga, A. (2021). Towards generalisable hate speech detection: a
review on obstacles and solutions. PeerJ Computer Science, 7, e598.
Yunusova, S. M. (2021). Paraphrases related to the language of
advertising. International Journal of Linguistics, Literature and Culture, 7(4),
236-240. https://doi.org/10.21744/ijllc.v7n4.1766