A Window into Disciplinary Discourse:
Titles of Research Articles
Robin Lee Nagano
Institute of Modern Philology nyerobin@uni-miskolc.hu
Keywords: academic writing, disciplinary discourse, genre analy- sis, research articles, titles
1. Introduction
Academic writing, and more specifically writing for scholarly publication, is not an easy task. It involves learning the basics – expressing logical argument, using the proper register, structuring information, describing methodology, support- ing claims, proper citation, and so on. This learning may be partially formalized in academic writing classes, but much of it takes place out of the classroom. This is because a successful academic writer must go beyond these building blocks to shape the materials in a way that suits the target audience – fellow scholars, in most cases. Disciplines have different ways of working and different focal points and over time develop their own conventions for academic communica- tion.1
It takes time, experience, guidance and attention to discourse conventions to become competent in writing for publication in a particular discipline. We know that learning to function comfortably in an academic discipline requires exposure and also effort, and the initiation of novices is a drawn-out process.2 When a second language enters the picture, the situation becomes even more complex. A scholar may be initiated into the disciplinary community in one cul- ture and language, but cannot necessarily carry over all of those conventions when writing for peers in a second language. If a scholar moves into a new dis- cipline, he/she will need to acquire a new set of writing conventions. Thus, all scholarly writers need to acquire an understanding of disciplinary conventions
1 Ken HYLAND, Disciplinary Discourses: Social Interactions in Academic Writing, (Harlow: Pearson Education, 2000); Tony BECHER and Paul R. TROWLER, Academic Tribes and Territories, (Buckingham:
Open University Press, 20012)
2 See e.g. Dacia DRESSEN-HAMMOUDA, “From novice to disciplinary expert: Disciplinary identity and genre mastery”, English for Specific Purposes, 27(2008) Nr. 2. 233–252.
in writing for publication. One key to developing skills is awareness of dis- course conventions.3
2. Why titles?
This paper considers one very small but significant aspect of a research paper – its title. Titles are important because they represent the full text. Their role is becoming even stronger as a result of digital databases and on-line resources.
In earlier days, the title was more of a label used for easy reference to an article in a journal. The article itself was in a printed volume held in the hand; it was easy to flip through it, scan the abstract (if even available), note the headings, check out the figures and tables, glance through the names in the reference list, read a few lines here and there, skim for key words, and make a decision on whether the article was relevant and worth reading.
These days the title most often appears detached from the article (in a Ta- ble of Contents, for instance, or as a result from a database search). While the name of the author(s) and the title of the journal provide some information, the main source of information for the potential reader is the title of article. There may be dozens – hundreds – thousands! – of articles to choose among, and thus the potential reader must go through a decision-making process each time:
check the box to select it or not? click on the article or go on to the next? In most cases going on leads to an abstract, which provides further information, based upon which the potential reader decides whether or not to access the full text.
As the first step in this process, the title is, if anything, growing in importance.
The importance of research article titles has been recognized by scholars, leading to a number of studies in different disciplines and focusing on various aspects of titles. The study of ‘titology’ is often attributed to Genette,4 who fo- cused mainly on book titles. Articles comparing titles of journal articles across disciplines.5 Numerous others analyze titles in one discipline6 or even one
3 HYLAND, Disciplinary Discourses, op.cit.
4 Gerard GENETTE, “Structure and function of the title in literature”, Critical Enquiry, 14(1988) Nr. 4.
692–720.
5 Madeline HAGGAN, “Research paper titles in literature, linguistics and science: dimensions of at- traction”, Journal of Pragmatics, 36(2004) 293–331.; Viviana SOLER, “Writing titles in science: An exploratory study”, English for Specific Purposes, 26(2007) 90–102.; Kingsley Richard APPIAH and Christopher ANKOMAH and Harrison Yaw OSEI and TimothyHATTOH-AHIADUVOR, “Structural Organi- sation of Research Article Titles: A Comparative Study of Titles of Business, Gynaecology and Law”, Advances in Language and Literary Studies, 10(2019) Nr. 3. 145–154.
6 E.g. Sara GESUATO, “Encoding of information in titles: Academic practices across four genres in linguistics”, in Ecolingua. The Role of E-corpora in Translation and Language Learning, ed. Christo- pher TAYLOR, (Trieste: Edizioni Universita di Trieste, 2008) 127–157.; Mary E. KERANS and Anne
journal. There is also interest in tracing title features in huge databases encom- passing hundreds of journals, such as Lewison & Hartley,7 who looked at title length and the use of colons, or Ball, who investigated the use of question marks.8 Some studies are diachronic, tracing changes in title features over time while other studies attempt to determine whether title features have any effect on citation rates and yet others investigate the influence of the first language on titles written in a second language.
In all cases, the primary purposes of the title are similar: to represent the full text, to inform potential readers, and also to attract potential readers.9 How- ever, titles can be framed differently in different discourse communities to ful- fill these purposes. This paper attempts to demonstrate some of the ways that this occurs.
3. The data
The data presented in this paper is part of a corpus of titles representing eight disciplines chosen to represent hard sciences and soft sciences, as well as more theoretical and more applied areas. The four soft sciences are economics, edu- cation, history, and sociology; the four hard sciences are botany, fluid engineer- ing, geology, and medicine. For the entire corpus titles were taken from 100 articles in four leading international general journals (mainly issued in 2006), making 400 titles per discipline, for a 3,200-title corpus, which was analyzed mainly by hand.10 For this paper, however, I have chosen a set of 100 titles from one journal of the four represented in the corpus (I randomly chose the third journal for each discipline). Based on the characteristics of each set of titles, I have selected a few titles that demonstrate typical features of that set and will use these to introduce features such as average length (in words) of the titles, common structures and styles, and lexical aspects such as word frequency and collocation.
MURRAY and Sergi SABATÈ, “Content and Phrasing in Titles of Original Research and Review Articles in 2015: Range of Practice in Four Clinical Journals”, Publications, 4(2016) Nr. 2. article 11.
7 Grant LEWISON and James HARTLEY, “What’s in a title? Numbers of words and presence of colons”, Scientometrics, 63(2005) Nr. 2. 341–356.
8 Rafael BALL, “Scholarly communication in transition: The use of question marks in the titles of scientific articles in medicine, life sciences and physics 1966–2005”, Scientometrics, 79(2009) Nr.
3. 667–679.
9 GENETTE, Structure and function of the title in literature, op.cit.; Christiane NORD Einführung in das funktionale Übersetzen: Am Beispiel von Titeln und Überschriften, (Tübingen: Francke, 1993)
10 For details see: Robin Lee NAGANO, “Research Article Titles and Disciplinary Conventions: A Cor- pus Study of Eight Disciplines”, Journal of Academic Writing, 5(2015) Nr. 1. 133–144.
4. Typical features in titles from eight disciplines
In this section I introduce titles that demonstrate typical characteristics of the 100-title samples and briefly discuss these features for each of the investigated disciplines.
4.1 Botany (sample from the journal Planta)
(1a) Molecular cloning and characterization of phosphorus starvation responsive genes in common bean (Phaseolus vulgaris L.)
(1b) Water deficits accelerate ripening and induce changes in gene expression regulating flavonoid biosynthesis in grape berries
(1c) The isochorismate pathway is negatively regulated by salicylic acid signaling in O3-exposed Arabidopsis
Title 1a contains both the common name and biological name of the plant being studied, which appears to be a common practice. Other times just the common name appears, as in Title 1b. The exception is Arabidopsis (thaliana), “a small flowering plant that is widely used as a model organism in plant biology”11, and which appears in 20 of the titles, including Title 1c.
In structure, Title 1a is essentially a complex noun phrase, both pre- and post-modified. In contrast, the other two titles are essentially full sentences, containing finite verbs (active in 1b: accelerate, induce; passive in 1c: is regu- lated). Such titles are most commonly found in life sciences;12 in this sample, 42 titles were of this type. Full-sentence titles tend to be somewhat longer: an av- erage of 16.9 words, while the noun-phrase type has an average of 14.8 words;
the overall average for the botany data set is 15.7 words. All three sample titles are single-unit titles, which is typical of this data set. Only 11 of the titles con- tained two units.
4.2 Fluid engineering (Journal of Fluids and Structures)
(2a) Effects of actuator nonlinearity on aeroelastic characteristics of a control fin (2b) Flow around two side-by-side closely spaced circular cylinders
(2c) Vortex shedding and aerodynamic forces on a circular cylinder in linear shear flow at subcritical Reynolds number
The average length of the titles in the sample is 12.1 words, and the majority of titles consist of one unit (10 titles have two units). The vast majority are modi- fied noun phrases, as seen in Titles 2a–2c. It is typical to have multiple preposi- tional phrases giving the description of the phenomenon being investigated and its parameters. Analysis of the preposition ‘on’ is rather revealing: it appears
11 “About Arabidopsis”. The Arabidopsis Information Resource. Online: arabidopsis.org
12 HAGGAN, Research paper titles, op.cit.
most often as a collocation of effect(s)/influence (as in Title 2b; appears in 8 titles); collocates with force(s)/impact/loading (as in 2c; 6); is used as a syno- nym of ‘about’ either within in a title (3 times) or as its initial word (3 titles);
appears in other collocations (‘based on’, etc.); and only once does it indicate location.
4.3 Geology (International Geology Review)
(3a) High- and Ultrahigh-Pressure Metamorphism in the North Qaidam and So- uth Altyn Terranes, Western China
(3b) Early Orogenic History of the Eastern Himalayas: Compositional Studies of Paleogene Sandstones from Assam, Northeast India
(3c) Pressures and Temperatures of Ultrahigh-Pressure Metamorphism: Implica- tions for UHP Tectonics and H2O in Subducting Slabs
The average length of titles is the sample is 16.1 words. Certainly one factor characteristic of geology contributes to this length: the location being studied is necessary information for the title. Ninety titles include geographical locati- ons (as do Titles 3a and 3b), ranging from broad descriptions (‘Caribbean’) to 3-level descriptions (usually geological feature, region, nation). The remaining ten papers focus on a general model, as in Title 3c. Nearly half of the sample titles (47) have two units, as in Titles 3b and 3c.
The lexical items appearing in titles are generally minerals, formations, landforms, geological eras or processes. Interestingly, several strategies are used to suggest that the researcher draws generalized conclusions from specific cases. One is to present a generalized topic, as in Title 3b. Another is to lean on lexical items such as ‘implication(s)’ (used in 12 titles); ‘evidence from’ (6); ‘in- terpretation’ (3); ‘insights’ (2); and ‘as inferred from’ (1). A less common claim is that of novelty, using the words ‘new’ (5) and ‘first’ (1).
4.4. Medicine (New England Journal of Medicine)
(4a) Single-Dose Azithromycin for the Treatment of Cholera in Adults (4b) Long-Term Outcome of Renal Transplantation from Older Donors (4c) Cyclophosphamide versus Placebo in Scleroderma Lung Disease
Title 4a is an example of the most typical pattern: medication/treatment for condition in patient group. Prepositional phrases are important in the structure of medical titles. A few titles add focus terms such as ‘Safety and efficacy of’,
‘Feasibility of’, but most title designers seem to feel that this is unnecessary (aiming for feasibility, efficacy, and safety would presumably be aims of most treatments). Other focus words are ‘outcome(s)’ (as in Title 4b), ‘effect(s) of’
and ‘trends in’, but these are rather infrequent. Perhaps this is a factor in the relatively low average word count of 9.7 words. Other space-saving techniques
are replacing ‘A comparison of x and y in…’ with ‘X versus y in…’ (Title 4c), and
‘Association of x with y’ with ‘x and y’.
No full-sentence titles appear in this general journal; although an earlier study of mine13 found that they are rare in the general journal corpus but com- mon in a title sample from specialist medical journals.14
All titles in the sample have one unit. This is clearly the result of editorial policy, as the other journals in the full corpus include two-unit titles,15 and BMJ requires a second unit with information on the method or type of study.16 In contrast, in this sample relatively few titles (5) give explicit information on trial design. A sample of titles for the same journals for articles published in 2015 revealed the same trends as in titles from a decade earlier,17 despite this going against the recommendations of the International Committee of Medical Jour- nal Editors.18
4.5. Economics (Review of Economic Studies)
(5a) Investment under Uncertainty in Dynamic Conflicts (5b) Measuring Peer Effects on Youth Smoking Behaviour (5c) Information Markets and the Comovement of Asset Prices
The average length of the sample titles is 7 words. The structure is most often a single unit (77%). The topic is often given further context through preposi- tional phrases, as in Title 5a. Alternatives to a noun as the initial word are also found: -ing forms (as in 5b) or prepositions such as ‘On’. There are relatively few occurrences of focus phrases such as “effect of x on y” or “the relationship between x and y”. Instead, as in 5c, the coordinate ‘and’ is often used (it is the most frequent word of all!). While the reader assumes there is some relation- ship between the two (or sometimes three) factors, this is not specified,
13 Robin Lee NAGANO, “Titles of biomedical articles: Data and a debate”, Porta Lingua, (2008) 115–
123.
14 This observation has been confirmed by KERANS et al.,Content and Phrasing in Titles of Original Research, op.cit.
15 NAGANO, Titles of biomedical articles, op.cit.
16 “The title should be informative and, for research papers, a subtitle with the study design (for example, „a phase III clinical trial” or „a systematic review and meta-analysis”).” Article Types at BMJ. Online: www.bmj.com / About BMJ / Article types. – December 2019.
17 KERANS et al.,Content and Phrasing in Titles of Original Research, op.cit.
18 “The title provides a distilled description of the complete article and should include information that, along with the abstract, will make electronic retrieval of the article sensitive and specific. Re- porting guidelines recommend and some journals require that information about the study design be a part of the title (particularly important for randomized trials and systematic reviews and meta- analyses).” – “Preparing a Manuscript for Submission to a Medical Journal”, Online: www.icmje.org / Recommendations – December 2019.
possibly leading the reader of the title to begin thinking of potential connec- tions or cause-effect relationships.
4.6 Education (Journal of Educational Research)
(6a) Factors Related to Instructors’ Willingness to Participate in Distance Educa- tion
(6b) Exploring Teacher Perceptions of Small Boys in Kindergarten
(6c) Supplemental Instruction in Early Reading: Does It Matter for Struggling Readers?
The average title length is 10 words and 68 of the titles are single units. While titles such as Title 6a are common, a wide variety of structures appear, includ- ing -ing forms (as in 6b); 20 titles begin with an -ing form that can be inter- preted as ‘how to’, reflecting an emphasis on taking action. Another strategy that appears is two-unit questions where one unit is a question (as in 6c). Titles units often contain ‘effects of’ (18 instances) or ‘impact of (8), indicating an ori- entation towards cause-effect (problem-solution) approaches. Coordinates are sometimes used to juxtapose two unmodified topics to create the impression of a relationship between them.
4.7. History (History, the Journal of the Historical Association)
(7a) Of Spies, Refugees and Hostile Propaganda: How Austria dealt with the Hun- garian Crisis of 1956
(7b) ‘On Account of their Disreputable Characters’: Parish-Assisted Emigration from Rural England, 1834–1860
(7c) The Politics of Sir Thomas Fairfax Reassessed
The average number of words is 12.3, and 31 of the titles consist of a single unit. Predictably, names of places, people, and events are common, along with periods or dates. The high proportion of multi-unit titles gives the opportunity to use a variety of structures: prepositions (Title 7a), wh- statements (2nd unit in 7a), gerunds, questions, and quotations (7b) and questions appear. One in- teresting lexical feature indicates coming back to topics with a new perspective, as in 7c: the sample contains revisited (3), reassessed, a reappraisal, and recon- sidered.
4.8 Sociology (American Journal of Sociology)
(8a) Black-White Wage Inequality, Employment Rates, and Incarceration (8b) Gendering the Job: Networks and Recruitment at a Call Center
(8c) Pulled, Pushed, and Persuaded: Explaining Women’s Mobilization into the Salvadoran Guerrilla Army
The average title length was 11.7 words, and the sample contained 27 single- unit titles. A variety of structures appears, including the coordinate ‘and’ used to indicate relationship (Title 8a). Title units that are not especially informative but that may attract readers’ attention are quite common, especially as the first unit of a two-unit title (8b, 8c). The alliteration in 8c is just an example of the many rather creative devices used to appeal to readers. Whether such catchy titles actually attract readers (and citations) is still a debated question.
Conclusions
I hope this brief survey of titles of research articles in different disciplines has demonstrated that conventions for meeting the purposes of a title vary by dis- cipline to some degree. Generally speaking, the hard sciences show less varia- tion in structure and style compared to the soft sciences (with the exception of botany’s full-sentence option). The features discussed here – title length, num- ber of units, title structure, collocations, etc. – are easy to access. Undergraduate students might benefit from a vocabulary list made up of words appearing in titles; those beginning to read the literature might look at structure; writers might want to carry out a more detailed analysis of journal titles in their spe- cialized field.
It is important to point out that one journal cannot represent a discipline, or even the category of international general journal within a discipline. The journal representing economics in this study is the most ‘hard-science-like’ of the four journals I studied, i.e. it has the highest proportion of single-unit titles and of titles initiated with a noun, as well as the lowest number of title units containing questions and the lowest average word count per unit.19 For a fuller understanding of title conventions, it is important to observe titles from a num- ber of publications. As noted earlier, general journals and specialist journals may differ. If a researcher is targeting a particular journal for publication, it would be worthwhile analyzing the titles from a few recent issues before final- izing a title for the article to be submitted. Teachers who wish to aid their stu- dents’ initiation into a discipline may find working with titles an effective way to open a window into the discourse community and the world of reading and writing about research in English.
19 NAGANO, Research Article Titles and Disciplinary Conventions, op.cit.
Digital Object Identifier (DOI) codes of sample titles
title DOI title DOI
1a 10.1007/s00425-007-0603-2 5a 10.1111/j.1467-937X.2006.0385.x 1b 10.1007/s00425-007-0598-8 5b 10.1111/j.1467-937X.2007.00448.x 1c 10.1007/s00425-007-0556-5 5c 10.1111/j.1467-937X.2006.00397.x 2a 10.1016/j.jfluidstructs.2007.04.003 6a 10.3200/JOER.99.2.109-115 2b 10.1016/j.jfluidstructs.2006.12.002 6b 10.3200/JOER.100.6.331-335 2c 10.1016/j.jfluidstructs.2006.11.004 6c 10.3200/JOER.99.2.99-108 3a 10.2747/0020-6814.49.11.969 7a 10.1111/j.1468-229X.2006.00359.x 3b 10.2747/0020-6814.49.9.798 7b 10.1111/1468-229X.00281 3c 10.2747/0020-6814.48.12.1053 7c 10.1111/j.1468-229X.2005.00347.x
4a 10.1056/NEJMoa054493 8a 10.1086/432780
4b 10.1056/NEJMoa052891 8b 10.1086/497257
4c 10.1056/NEJMoa055120 8c 10.1086/502690
My major publications on this topic:
Robin Lee NAGANO, “Titles of biomedical articles: Data and a debate”, Porta Lingua, (2008) 115–123.
Robin Lee NAGANO, “Lexical comparison of journal article titles in soft disciplines”, Porta Lingua (2009) 111–117.
Robin Lee NAGANO, “Semi-technical vocabulary: The lexis of research article titles”, Porta Lingua (2010) 100–111.
Robin Lee NAGANO, “Research Article Titles and Disciplinary Conventions: A Corpus Study of Eight Disciplines”, Journal of Academic Writing, 5(2015) Nr. 1. 133–144.
