LINGUIST List 30.1649

Tue Apr 16 2019

Review: Forensic Linguistics; Text/Corpus Linguistics: Olsson (2018)

Editor for this issue: Jeremy Coburn <>

***************** LINGUIST List Support *****************

Fund Drive 2019
29 years of LINGUIST List!
Please support the LL editors and operation with a donation at:

Date: 01-Feb-2019
From: Brett Drury <>
Subject: More Wordcrime
E-mail this message to a friend

Discuss this message

Book announced at

AUTHOR: John Olsson
TITLE: More Wordcrime
SUBTITLE: Solving Crime With Linguistics
PUBLISHER: Bloomsbury Publishing (formerly The Continuum International Publishing Group)
YEAR: 2018

REVIEWER: Brett Mylo Drury

Forensic Linguistics has recently emerged into the public consciousnesses through Netflix’s dramatization of Ted Kaczynski, the Unabomber, who unwittingly revealed his identity through the publication of his anti-technology thesis. His use of the phrase of “can’t eat your cake and have it too” was identified by his brother who informed the authorities. In conjunction with this surge of interest from the general public, there has been an increased interest from the research community, in particular there is a centre of excellence at Aston University located in the UK. In addition, Belinda Maia, from the University of Porto has over the years arranged some excellent Forensic Linguistics workshops. This research has in turn been used in criminal and civil trials in various jurisdictions. Against this background John Olsson has published More WordCrime, which is his second attempt to publicize Forensic Linguistics to the public at large.

More WordCrime is divided into four parts: “Toolkit”, “Confronting Authority”, “The Authority To Confront”, and “Life in Forensic Linguistics”. Each of these of parts contains one or more chapters which is loosely related around the the theme of its containing part. In total the book contains 22 chapters, most of which are 20 pages or shorter. The large number of chapters and their relatively short length is both a source of strength and a weakness of the book. The short length of the chapter ensures that the casual reader does not got bogged down in technical detail, but the motivated reader will feel short changed about the lack of depth of each chapter. Olsson does provide some suggestions for extra resources, but this is not for every chapter, and the number of suggested resources is limited. In addition Olsson tends to veer off topic into politics, which detracts from the central theme of the book. The most obvious example of this is in Chapter 2: “The Linguistic Tragedy of Hillsbrough”, where the majority of the chapter is devoted to authoritarian power structures in British Society. These minor quibbles aside, this book is a well written introduction to Forensic Linguistics for the general public.

The structure of this review will be centered around each part of the book in order of publication and the chapters it contains.


The Toolkit Part of the book contains a single chapter: How To Do Forensic Linguistics. This chapter breaks down Forensic Linguistics into constituent parts as well as discussing the limitations of Forensic Linguistics, such as the fact that Forensic Linguistics cannot determine if the authors are telling the truth. The chapter also describes the different role of the forensic linguist as an expert and the solicitor (lawyer) in a criminal trial. The remainder of the chapter describes the process of document analysis as conducted by the forensic linguist, as well as the linguistic phenomena that the expert is looking for, which are: Spelling, Grammar, Lexicon, and Idiom. Spelling is the spelling and punctuation habits of the author nder study, such as misspellings as well as the use of certain recurring spelling patterns such as using an ‘s’ in words such as ‘organise’ rather than the more common ‘organize’. Grammar refers to the sequence of words, and the rules of grammar that are uniquely interpreted by the author under study. Lexicon refers to the word choice by the author, whereas Idiom refers to unique multi-word expressions such as the aforementioned “can’t eat your cake and have it too” which can be traced to the author. Olsson concludes that the forensic linguist can’t definitively attribute authorship, but only offer an opinion.

Confronting Authority

Confronting authority contains five chapters: ''The Linguistic Tragedy of Hillsborough'', ''A pink handled kitchen devil knife and other fabrications'', ''I didn''t have a gun'', ''All quiet at the endz'' and ''Wars and words''.

''The Linguistic Tragedy of Hillsborough'' covers a sporting event, which is burnt into the conscienceless of the British Public, where there were 96 fatalities and 766 injuries which was due to the negligence of the Merseyside Police. The Police covered up their negligence, and with the aid of The Sun newspaper attempted to attribute blame to the fans themselves.

The beginning of the chapter is a brief discussion of the general role of Forensic Linguistics in determining what is happening when people in power cover up a tragedy or an abuse of power. Olsson describes a presentation given by his students from the University of Bangor to lawyers about the ability of Forensic Linguistics to assist in criminal investigations. Olsson makes some statements about the inability of computers to match the language analysis abilities of humans, a statement which this reviewer cannot agree with.

The majority of the chapter is concerned with the events that occurred before, during and after the tragedy as well as the subsequent cover up by various political actors. Olsson provides some example statements made by the aforementioned political actors as well as providing some evidence of collusion between high ranking police officers when making statements to the inquiry. The chapter rounds off with a discussion of the power of state and its agents,and its relationship with the individual.

This chapter is by far the longest of the book and probably the weakest, as it veers off topic and descends into a political diatribe.

''A pink handled kitchen devil knife and other fabrications'' is also concerned with evidence fabrication. Olsson describes various techniques how police officers collude in recording and making statements. And how that there is a pressure from the attitudes of senior ranking officers to implicitly force police officers to produce similar sounding statements. He also documented some experiments that show human recall of actual words used to be poor after a short period of time. He also stated that lexical dense phrases are unlikely to be used in spoken language. This brings us to the “A pink handled kitchen devil knife” chapter title. This statement was allegedly used by a defendant when he admitted to a stabbing. Along with poor recall of speech, and the very low probability of a “lexically dense” phrase being used in everyday speech, it is very likely that this phrase was fabricated to gain a conviction. Olsson also discussed copying and pasting chunks of text from one statement to another, and provided an obvious example where two police officers colluded to such a degree that their IDs and names where used interchangeably in both of their statements. Olsson, however concludes that judges are indifferent when they come across obviously colluded statements from police officers.

''I didn’t have a gun” is a short and concise chapter about defendant statement fabrication by police. This chapter described an occurrence where a youth was forced to sign a confession that he did not write or agree to. The chapter picks apart the contradiction of vague statements such as “towards the beginning of August” and the specific “a Wednesday towards the beginning of August”. The vagueness in the statement is used to demonstrate untrustworthiness. The chapter dissects the alleged confession with obvious examples of phrases which the defendant is unlikely to use. And finally the author concludes that police use questions that elicit a denial as a way of framing language to infer guilt.

''All quiet at the endz'', describes an attempt by the police to infer that performers in a rap video had committed criminal offences that they describe in their rhymes. This chapter at least demonstrates the limitations of Forensic Linguistics as this was not possible.

The Confronting Authority part concludes with ''War and Words'', which describes the use of misleading language in the recent Brexit referendum where the UK voted to leave the European Union. Like the first chapter in this part, it shows Olsson at his weakest where the chapter descends into a politics essay. The linguistic part of this chapter focuses on manipulative language such as framing. The chapter signs off with a barb at the people Olsson assumes voted to leave by declaring that there are no “pure British people”.

The Authority To Confront

This part of the book contains six chapters: “Not a case of plagiarism”, “How old? What gender”, “Alarm and Distress”, “The prosecutor of the ICC v the President of Kenya”, “The Facebook murder” and “The sting”.

‘Not a case of plagiarism” centres upon a case of plagiarism within a university where a student had suddenly increased her grade from 60% to 80% from her midterms essay to her end of course essay, and the University in question took the view that this increase in performance could only be due to cheating. Her situation was compounded by her posting a request for help on the Internet. The chapter describes Olsson’s contempt for essay mills who in his opinion produce low quality essays and he goes on to describe the characteristics of an essay produced by an essay mill. These characteristics include: nonsensical references and poor quality content. Olsson compared her midterm essay with her final essay and compared a number of linguistic indicators including hyphenated words and the method of referencing. This indicated that the mid-term and final essays were written by the same person, and therefore the final essay was not plagiarised.

“How old? What gender” is concerned about a sexual offences trial where a text message was purported to be sent by females under the age of 16. The text message stated that they were willing to partake in sexual activity with the defendant. A second text message was sent from an adult stating that they could procure minors for sex. The chapter describes social linguistic variation, where the author changes the manner in which they express themselves in written form. This characteristic is more prevalent in females under the age of 16. In the case of the text messages Olsson concluded that the text message allegedly written by females under the age of 16 was likely to be written by an adult.

“Alarm and Distress” is a short chapter that describes a series of threatening letters sent to a business woman. The business woman had a suspicion of a possible suspect, a businessman that she had romantically rejected. The use of various specifics phrases such as “D’You” in other correspondence authored by the man indicated that her initial hunch was correct.

The prosecutor of the “ICC v the President of Kenya” chapter is concerned with the indictment of President Kenyatta for post election violence in 2008. In this chapter Olsson was tasked with discovering if the witness statements were the product of collusion or were written individually. Olsson found evidence of “evidence-tampering, collusion between witnesses and mass plagiarism.” He concluded that there was one author behind all of the witness statements. In part due to Olsson’s work the charges were dropped against Kenyatta. The evidence that Olsson could show was limited, but he manages to show some heavily redacted documents and their similarities.

“The Facebook murders” describes the role of Facebook messages in assisting the investigation of a murder of a female teenager. The case involves a deception by 33 year old convicted rapist who arranged a date posing as a fellow teenager. The 33 year old sent messages over Facebook posing as a teenager, as well as the father of the teenager. Olsson found that the messages sent by the fictional teenager and the 33 year old shared linguistic features. After murdering the teenager the 33 year old sent messages to her colleague. The language in these messages did not conform to the expected idioms of a young person. This research helped in the conviction of the 33 year old.

“The Sting” chapter describes a businessman who was manipulated by a female model to unwittingly smuggle a suitcase full of cocaine. The businessman never met the model, but communicated with her through the Internet. The businessman was fired from his job and he was prosecuted for drug trafficking based upon 3 emails that were purportedly sent by the businessman, indicating that he knew that the suitcase contained narcotics. Olsson concluded that these emails had not been authored by the businessman, but likely by a Spanish speaker with knowledge of the drug trade.

Life in Forensic Linguistics

This part is the largest of the book as it contains ten chapters: “Nothing is not important”, “When authorship is not authorship”, “A letter for Mrs Joe”, “The strange pose of Mrs Mottle”, “The love letters of Dr X”, “The invisible Bronski”, “Dissing the opposition”, “The concrete tomb”, “A particularly unpleasant man”, and the “ The mysterious Mr Erdnase”.

Nothing is not important” discusses the role of unique or non-standard punctuation and its role assisting the forensic linguist. The chapter provides a number of examples of the aforementioned phenomena, and as fitting for the modern age, Olsson also provides some examples of emjoi idioms. He ends with the chapter by stating that he ensures that his students transcribe documents so that they improve their observations and are able to identify these non-word idioms.

“When authorship is not authorship”, is concerned with wills, and the specific problems that they create for the forensic linguist. Wills are an expression of intent of the testator; however the testator may not be the author of the will because a 3rd party could write the will. This unique characteristic of a will renders authorship analysis ineffective for determining if a will represents the true wishes of the testator.

“A letter for Mrs Joe” addresses the authorship of a poison pen letter sent to the wife of a particularly staid couple which accused the husband of womanizing. The author claimed that “she” was sleeping with the husband, and had only just become aware that he was married. The husband managed to gather some specimens of writing from a number of his colleagues at his workplace. Olsson found a number of idioms that were unique not only to the poison pen letter, but to the writing specimens from one individual. The evidence however was not conclusive. Although the individual was not confronted, they confessed to writing the letter to a third party.

“The strange prose of Mrs Mottle” chapter concerns the authorship of inflammatory letters to the board of trustees of a charity that was set-up to protect a certain species of wildfowl. One of the trustees was accused of misappropriating funds. He wrote a series of letters complaining about his treatment. His “wife” wrote a number of more colourful letters to the board. Olsson performed some analysis on the letters from both the wife and her husband. He examined the number of common expressions that are more than six words long, the register and the orthography of the letters. He found three common expressions, and deduced that this would be unlikely if the letters were written independently. In addition the register of the letters was similar because both authors used “pompous and archaic expressions”. Finally, the use of punctuation was similar in both specimens of letters. Olsson concluded that the letters were written by the same person, a fact reinforced by the fact that the wife had died several years before the letters were written.

“The love letters of Dr X” chapter described an incident where a female Pakistani doctor practicing in the UK had been accused of writing love letters to a elderly male patient. In this case Olsson looked at idioms. Idioms can be good determiner to separate native and non-native speakers of English. In this case the love letters showed evidence that the author had a strong familiarity with English idioms. In addition the standard of writing and comprehension displayed by the doctor was weak, which was in contradiction to the level of English demonstrated by the author of the love letter. Olsson concluded that the doctor was unlikely to be the author of the love letters.

“The invisible Bronski” chapter follows the format of the previous chapters in this part where distasteful communications are targeted at individuals. In this case the malevolent communications took the form of reviews of products of a specific businesswoman. The reviews contained accusations of child abuse. In conjunction an identifiable person started posting critical reviews, but these were written in a manner so that the reviews avoided libel. The problem in this case was to determine if the identifiable author was also the author of the distasteful reviews. Again the chapter describes some analysis performed by Olsson which included the analysis of redundant language. Olsson concluded that the identifiable author authored both the distasteful reviews and the non-libel ones.

“Dissing the opposition” is a short chapter that discussed the positing of fake negative product reviews of a manufacturer by a competitor. Olsson compared the fake reviews with the text of websites of direct competitors. He found that one competitor’s website matched the writing style of the fake negative reviews.

“The concrete tomb” chapter describes an incident where a native French speaker (Christophe) had gone missing for two years. His last known communication was an email he sent to an acquaintance that stated he was “going on holiday” . After these two years had passed, a male (Sebastien) confessed to murdering Christophe as an accomplice to another male (Dominik). The authorship challenge was to determine if Christophe’s last communication was written by him or another author. The chapter discusses the unique writing habits of French speakers, in particular the use of accents. The chapter presents a number of examples of writing from both Christophe and Sebastien. Olsson concludes that it was likely that Dominik was the author of the email.

“A particularly unpleasant man” is centred around a murder case and an alibi for the alleged murderer that was based upon mobile phone messages sent from the victim's phone. The chapter follows a familiar path where an analysis is conducted, and Olsson concludes that the alleged murderer sent the messages after he had committed the murder.

The final chapter discusses the authorship a famous magic book, Expert. There were a number of suspected authors. Despite the analysis by Olsson, he was unable to conclude conclusively who was the author.


This is a popular book which linguists may find lacking a depth; however it is an excellent introduction to the field for the beginner or casual reader. It is a large number of chapters, some having 10 pages or less. This keeps the pace of the book fast, but contributes to the book’s lack of depth. Many of these chapters are similar and consequently the book feels repetitive. However in general, despite these criticisms, it is an excellent popular book discussing the application of a academic subject.


Brett is currently the Head of Research at Scicrop, a start up located in Sao Paulo, Brazil. He holds a PhD in Computer Science, and a qualifying undergraduate law degree. He is interested on the quantifying impact of the language of the law upon the justice system.

Page Updated: 16-Apr-2019