Differences in Journal Paper Titles between Natural Sciences and Humanities: A Corpus-based Analysis

4.1.1 The Length of Titles

In general, the majority of international academic journals have guidelines regarding the title writing, especially the word count of article titles. For example, the London Mathematical Society demands fewer than 12 words, and Journal of the National Cancer Institute requires no more than 14 words (Liu 2017). These differences in guideline adherence and possibly other pressures on title create practical problems for authors (Kerans et al. 2020). Though it is a rigid requirement for each different journal, the discipline also plays an important role in determining how scholars name their RAs.

Hyland and Zou (2022) pointed that title length is important for the retrieval and the eventual citation of articles. The longer the title is, the more information it reveals to its readers. After a careful examination of titles in the corpus, we have noticed some interesting features about the length of titles in natural sciences and humanities, both similarities and differences could be found.

As shown in Table 2, the average length of natural science titles is longer than that of humanities, which is consistent with previous studies (Haggan 2004). Moreover, the longest title (29 words) and the shortest one (1 word) are both from humanities. For instance, “Sufficientarianism” is the shortest one in humanities, while “Nonconjugate Quantum Subsystems” is the shortest in natural sciences, referring to a philosophical and a physical term respectively. However, neither of these titles gives enough information to their readers, thus are not being viewed as a good title, and the number of such titles are not quite big in our corpus.

Xiang and Li (2020) found that the average length of RA titles in linguistics and literature has increased. Our study also shows that the average length of humanities titles is shorter than those in natural sciences, but we further explore the reasons from the perspective of disciplinary research focus and writing norms. As Milojević (2017) pointed out, discipline is the strongest determinant for the average length of titles and the occurrence of different forms of titles. It suggests that authors belonging to different disciplines need to comply with the norms set in their respective fields. In natural sciences, the title is predominantly centered on phenomena, processes, or relationships in the physical world. This is largely because natural sciences are dedicated to exploring and understanding the objective aspects of our surroundings. For example, the title, “Temperature Robustness in Arabidopsis Circadian Clock Models Is Facilitated by Repressive Interactions, Autoregulation, and Three-Node Feedbacks”, reveals under what conditions (repressive interactions, autoregulation, three-node feedbacks) this biological phenomenon (temperature robustness) occurs. Such detailed information leads to a relatively lengthy title. In addition, titles in natural sciences often highlight the discovery, explanation, or application of scientific principles. In many cases, they are designed to communicate new findings or validate existing hypotheses, enabling fellow scientists to quickly recognize the significance of the research and its contribution to the field. The title “A Dichotomy in Cross-Coupling Site Selectivity in a Dihalogenated Heteroarene: Influence of Mononuclear Pd, Pd Clusters, and Pd Nanoparticles – the Case for Exploiting Pd Catalyst Speciation” examplifies this perfectly. Through its precise terminology, it not only locates the study within catalysis niche, but also challenges prior assumptions and highlights a novel discovery, efficiently conveying the research value to peers.

On the other hand, researches in humanities center neither on experimental design nor fixed objective results (Xiang and Li 2020). As a result, titles in humanities are notably diverse and often more flexible compared to those in natural sciences, hence usually shorter. For example, in literary studies, a title such as “Who Bought Paris? Hope Mirrlees, the Hogarth Press, and the Circulation of Modernist Poetry” exemplifies the use of interrogative phrasing to foreground a research question. By directly imposing a question, this style of title not only captures attention and sparks curiosity but also explicitly signals the central inquiry of the study, the dynamics of poetry circulation through specific cultural channels. Metaphorical or allusive expressions are also frequently employed in humanities titles to add artistic and cultural connotations. A title like “A Wrinkle in Medieval Time: Ironing out Issues Regarding Race, Temporality, and the Early English” uses the metaphor of issues regarding race, temporality and the early English being a wrinkle in Medieval Time, conveying the idea that there are many serious problems about early English study. These diverse structural features in humanities titles reflect the nature of the humanities disciplines themselves, which focus on human experiences, cultures, thoughts, and interpretations. They aim to attract a wide range of readers by offering unique perspectives and inviting them to engage in in-depth thinking about various aspects of human life and society.

4.1.2 Grammatical Structures of Titles

The formulation of a title exerts a significant influence on the conveyance of the title’s information and serves as a manifestation of the author’s writing stance. Consequently, a comparative examination of the disparities in the title structures between natural sciences and humanities journals has the potential to disclose the impact of subject variables on the process of title composition. Following Haggan’s (2004) classification, the study focused on the five main grammatical structures of titles: nominal phrase (NP), compound phase (CP), verbal phrase (VP), prepositional phrase (PP) and full-sentence (FS). The distribution of different grammatical structures is shown in Table 3. The further chi-square test showed significant differences (χ²= 242.4, p < 0.05) in grammatical structures between RA titles in natural sciences and humanities.

In agreement with previous studies on RA titles (Haggan 2004; Soler 2007; Wang and Bai 2007; Xie 2020), our counts shows that both natural sciences and humanities do not prefer PP titles, accounting for 0.8 % and 0.6 % respectively. The most favorite structure in natural sciences goes to NP, up to 64.2 %, while in humanities, CP gains the most popularity, accounting for 58.4 %. A more detailed illustration of different structural patterns across six disciplines from natural sciences and humanities is presented in Figure 1.

Figure 1:

Structural pattern of titles across six disciplines.

As shown in Figure 1, the NP titles are most commonly used in natural sciences, and are observed in every discipline in our corpus. NP titles are noun-centered, free of sentence structure, which can simplify titles in the natural sciences, thus making them less redundant. For example, titles like “Experimental Study on the Conductivity of Nanomaterials” and “Continuous Flow Generation of Acylketene Intermediates via Nitrogen Extrusion” clearly convey that the research involves conducting experiments or undergoing specific procedures. This structure helps readers immediately recognize the methodology used in the research and the general area of focus. While in humanities, the NP titles are used to reveal the content of RA, as in titles like “The Gospel of Creativity” and “The Stoic Comedy of Elizabeth Bishop and Buster Keaton”. Besides, humanities writers tend to use multi-noun structures. We see a title like this, “Japan, the Ambiguous, and My Fragile, Complex and Evolving Self”. Only conjunctions and punctuation have been used to juxtapose five nouns, implying that they are somehow related. However, these titles can never been found in natural sciences, for it does not show any internal relationship among the nouns presented.

The CP title typically employs two-part structure connected by colons (most prevalent), dashes, or occasionally question marks, with periods being increasingly rare in contemporary practice. Compound headings attracted attention as early as the early 1980s. Dillion first discussed the use of colons in the RA titles, arguing that the colons in titles were the main characteristics of academic publications (cited in Liu 2017). As shown in Table 3, CP titles are most welcomed in humanities, and various compound titles are observed in our corpus. For example, the use of question marks in the title “Thales – The ‘First Philosopher’? A Troubled Chapter in the Historiography of Philosophy” is aimed to questioning purposes, activating curiosity of readers towards the research. Colons, dashes, and periods are also employed to structure relationship between title elements. For instance, “Assessing the Neoliberal Kunstlerroman:‘Creative’ Self-Realisation and the Art World in Michael Cunningham’s By Nightfall” uses a colon to separate its theoretical framework (neoliberal artist-novel) from its textual analysis (self-realization critique), while single quotation marks problematize the term ‘creative’, demonstrating how punctuation guides precise scholarly reading. Overall, by using different compound structure, titles in humanities offer a window into the rich tapestry of human experiences, cultures, and ideas, encouraging readers to engage in deep thinking and interpretation, and to view the world from diverse humanistic perspectives.

VP titles rank both the third in natural sciences and humanities. VP titles begin with a verb, followed by one or more noun phrases as an object, and may also use a prepositional phrase as a postmodifier. Such titles are not highly represented in all structures, as shown in the Table 3, with natural sciences accounting for only 5.6 % and the humanities 3.8 %. As Thompson (1993: 185) observed, whereas nominal groups combine grammatical and lexical items to encode multiple semantic functions (e.g. identity, qualities, classification, etc.), verbal groups are lexically streamlined, with their core meaning residing solely in the main verb, which express the Event. For example, titles like “Using a Smartphone Camera to Explore Ray Optics beyond the Thin Lens Equation” and “Exploiting Sodium Coordination in Alternating Monomer Sequences to Toughen Degradable Block Polyester Thermoplastic Elastomers” in natural sciences explicitly encode both the research methodology (smartphone-based experimentation, sodium coordination manipulation) and the scientific objective (extending optical theory, enhancing polymer toughness) within a single action-oriented syntactic frame. On the other hand, VP titles in humanities are typically prioritize conceptual engagement. Titles like “Recovering Franz Kafka’s Asbestos Factory” and “Revisiting the criticisms of rational choice theories” focus primarily on signaling the research domain (Kafka studies, political theory) and intellectual intervention (recovery, critique), often leaving methodological approaches implicit in the argumentation.

PP titles adopt an initial preposition followed by a nominal group. This structure aligns with Halliday and Martin (1993: 213) observation that prepositional phrases lack the logical Head-Modifier structure typical of nominal/verbal groups. Due to this grammatical constraint, where prepositions primarily denote relational rather than substantive meaning, PP titles struggle to encode self-contained research claims. Consequently, they remain statistically marginal in academic titling, accounting for merely 0.8 % of natural science journals and 0.6 % of humanities journals in our corpus. Despite its structural constraints, authors in natural sciences strategically leverage prepositions to achieve functional precision. Titles like “On levitation by blowing” and “Toward Comprehensive Exploration of the Physisorption Space in Porous Pseudomaterials Using an Iterative Mutation Search Algorithm” use directional prepositions (on/toward) to establish the study’s conceptual parameters, while the subsequent nominal groups encode methodological specifies (blowing techniques, iterative mutation search algorithm). This dual functionality allows scientific PP titles to compensate for their grammatical limitations. In contrast, the humanities employ PP titles like “On Sylvia Wynter and Feminist Theory” and “On Acknowledgments” in their more prototypical form. The preposition on here functions primarily as a topical marker, signaling the subject of inquiry without attempting to bundle methodological or theoretical complexity. It is the reflection of humanistic scholarship’s traditional emphasis on interpretive focus rather than procedural transparency.

FS headings are characterized by their SVO or SV structure, which includes both declarative and interrogative sentences. Full sentence construction allows researchers to present the general findings of their research both conclusively and synthetically in one sentence (Soler 2007: 98). A substantially larger proportion of this structural pattern is observable in the natural sciences as compared to the humanities. Consider the following representative examples extracted from natural sciences journals: “Tension Promotes Kinetochore-Microtubule Release by Aurora B Kinase” and “ZNF416 is a Pivotal Transcriptional Regulator of Fibroblast Mechanoactivation”. These titles effectively communicate highly precise details concerning the research scope, central focus, and methodological approach. This clarity facilitates seamless communication among researchers within the discipline, allowing them to swiftly gauge the pertinence of a given study to their own ongoing research efforts. In other words, such titles serve as a sort of shorthand that cuts through the complexity, enabling professionals to quickly identify whether a particular piece of research might hold value for them, thereby streamlining the process of knowledge dissemination and collaboration in the scientific community.

As Figure 1 further illustrates, FS structures mainly appear in biology, which is consistent with Soler’s study (2007). In our study, 65 out of 73 FS titles are from biology. In modern biology, there is a prevalent preference for verb-dominated expressions. The employment of the active voice, which serves to emphasize the key points, effectively conveys the author’s unwavering confidence in their research findings. FS titles are adept at stating and accentuating a distinct idea or conclusion, thereby facilitating clear communication within the scientific community. What’s more, being good at using tenses can also avoid misunderstandings when a particular idea is subsequently refuted or disproven (Zeng and Liu 2016). This is of particular significance in academic and scientific communication, where the accurate conveyance of information over time is crucial to maintaining the integrity and clarity of the discourse. Scholars from natural sciences are able to employ such titles to signify the temporal context of their research outcomes. However, it is essential to note that in situations where research findings are found to be inconsistent or contradictory, the inappropriate application of the present tense within titles has the potential to sow confusion within the academic community.

Conversely, in the field of humanities, there is a greater propensity to utilize interrogative forms to render headlines more engaging and thought-provoking. This interrogative mode of expression is an established and acceptable linguistic convention among members of the humanities speech community, as it serves to incite readers to engage in anticipatory contemplation. For example, such titles like “What is the true meaning of justice in a pluralistic society?” and “Why do we keep revisiting the classics in different eras?” are often seen in humanities. Nevertheless, such an expression holds relatively little significance within the natural sciences domain and is likely to be disregarded or excluded, given the distinct nature of scientific communication and the emphasis on objective, declarative statements.

4.2.1 High-Frequency Words

It is commonly acknowledged that high-frequency words have distinct stylistic specificity (Xiao and Zhang 2016). Natural sciences and humanities journals, belonging to different communities of practice, possess their own sets of terms and concepts. Certain common vocabulary items might carry discrepant definitions across different disciplines, and moreover, a multitude of newly coined words are exclusive to a particular field. The members within these communities continuously employ such specialized words, thereby fortifying the bonds among themselves and generating unique linguistic resources during their interactions. In light of this situation, crafting a title requires integrating subject-specific characteristics. This helps meet disciplinary norms, standardize vocabulary usage, and improve communication clarity within the academic community.

We used Concordant to search for the function and content words used in titles in natural sciences and humanities, and the distributions are shown in Figure 2 and Table 3. As the total number of words in two sub-corpora is not in parallel, we converted the original frequencies into frequencies per thousand words to eliminate the influence of corpus size. As we can see in Figure 2, the occurrence rate of words exhibits distinct disciplinary characteristics. The top 10 function words in our corpus include 7 prepositions (of, for, via, from, to, by, with), 2 articles (the, a), and 1 conjunction (and). Comparative data reveal that prepositions are more likely to appear in the titles of natural sciences. This is mainly because authors in this field need to express specific research methods, conditions, scopes, and results. For instance, four prepositions are used concurrently in “Development of a Concise and Robust Route to a Key Fragment of MCL-1 Inhibitors via Retrospective Defibrillation”, a title from natural sciences.

Figure 2:

Distribution of function words (frequency per 1,000 words).

Moreover, the reason why the occurrence of NP titles in natural sciences is significantly higher than those in humanities (as shown in Table 3) is partly due to prepositions facilitating the nominalization of titles through their functions of introduction, transformation, and marking (Xiao and Zhang 2016). In contrast, writers in humanities are more likely to use “and” to create a juxtaposition to explore the relationships between research contents, such as in “Ecological limits: Science, justice, policy, and the good life”.

Regarding the articles, “the” is used much more frequently in humanities than in natural sciences, while “a” appears more often in titles of natural science papers, which is consistent with Nagano’s (2015) study, who has found that hard and soft sciences are quite different in their patterns of direct article use. In English grammar, “the” is typically used before specific persons, books, phenomena, places, times, etc., which are the common subjects of study in the humanities. The definite article can be found in almost every title, like in “Naming Plantations: Toponyms and the Construction of the Plantation system in the English Atlantic”. On the other hand, “a” is used before an abstract or material noun with a modifier to emphasize a particular quality or class. For example, “a” modifies “Intermediate” and “Electrocatalyst” in “A Spectroscopically Observed Iron Nitrosyl Intermediate in the Reduction of Nitrate by a Surface-Conjugated Electrocatalyst” from natural sciences.

Table 4 presents a selection of high-frequency content words prevalent in both the natural sciences and the humanities. The significant disparity in the vocabulary utilized across these two fields strongly suggests that each discipline possesses its distinct set of language resources. Upon close examination of the corpus, it becomes evident that the majority of high-frequency words appearing in the titles of humanities journals are closely associated with temporal concepts (such as history, early, century, war). In contrast, the high-frequency words in natural science titles tend to embody the inherent nature of the discipline (like dynamics, catalyzed, quantum), or signify the research methodologies (for example, model, synthesis) and the objects under investigation (such as effect). This clear differentiation not only reflects the diverse subject matter but also the unique ways in which knowledge is constructed and communicated within each domain.

4.2.2 The Usage of Words of Different Parts of Speech

Besides the differences in terms of high-frequency words used in titles, we further investigate the words of different POS used in titles from these two domains. The result is shown in Table 5. A significant difference is found in POS usage between natural sciences and humanities titles (χ² = 30.13, p < 0.05), confirming the pattern is statistically reliable.

As shown in Table 5, the quantities of verbs and adjectives present in the titles of natural sciences are substantially greater than those in humanities. In academic arenas, the annual output of natural science papers is voluminous, dwarfing that of humanities papers by a significant margin. Concurrently, the retrieval of papers predominantly occurs via Internet search engines. This implies that the more keywords a title incorporates, the higher the likelihood of the paper being discovered by those with relevant interests. Within the context of journal papers, the interaction among natural science authors extends beyond mere academic exchange; it is also marked by intense competition. This competitive dynamic directly manifests in their strategic title construction, where authors employ multiple adjectives and verbs to precisely characterize their work. In the title “Development of a Concise and Robust Route to a Key Fragment of MCL-1 Inhibitors via Stereoselective Defluoroborylation”, the author uses three modifiers (concise, robust and stereoselective) simultaneously to outline both the methodological approach and the significance of the findings. In contrast, such a competition-fueled phenomenon is far less pronounced in the realm of humanities journal papers. Here, the focus seems to lie more on the conveyance of ideas in a relatively straightforward manner, perhaps due to the nature of the subject matter, which often centers on interpretations, analyses, and discussions that do not require the same level of highly specific terminological precision as in the natural sciences.

Firstly, with regard to verb usage, papers published in natural science journals exhibit a distinct pattern. Not only do they employ verbs to construct complete sentences as titles, but they also frequently resort to gerunds or the past tense of verbs to modify key or central words. Take, for instance, the title “Benchmarking Isotropic Hyperfine Coupling Constants Using (QTP) DFT Functionals and Coupled Cluster Theory”. In this heading, several verbs in such forms (benchmarking, coupling, using, coupled) are strategically incorporated. In contrast, scholars specializing in humanities also use gerunds when devising titles, as exemplified by “Reading Objects, Watching YouTube, Writing Biography, and Teaching Life Writing: Student Engagement When Learning Online During Covid-19”.

Secondly, titles in both natural sciences and humanities include a large number of nouns, with a similar frequencies per 1,000 words in our corpus (see Table 5). However, the usage of nouns present different pictures in two sub-corpora. For example, proper nouns surface more prevalently in the humanities domain. As we can see in the title “The Right to a Favor: International Scholarships, Clientelism, and the Class Politics of Merit in Post-Revolutionary Mexico” from humanities, multiple proper nouns (the Right to a Favor, International Scholarships, Clientelism, Class Politics, Post-Revolutionary Mexico) are featured. Conversely, the natural sciences lean more towards common nouns, which are often employed as modifiers. Consider the example “Time-resolved Electronic Sum-frequency Generation Spectroscopy with Fluorescence Suppression Using Optical Kerr Gating”. Here, five common nouns (sum-frequency, generation, spectroscopy, fluorescence, suppression) are utilized. This disparity indicates that authors in the humanities tend to focus on specific individuals or entities, while those in the natural sciences direct their attention to broader categories or groups of people or things.

Thirdly, titles in natural sciences employ a greater number of adjectives. This is primarily to enhance the modification of central terms and to convey precise information. For instance, in the title “Cytokine Responsive Networks in Human Colonic Epithelial Organoids Unveil a Molecular Classification of Inflammatory Bowel Disease”, the term “organoids” is meticulously modified by “human”, “colonic”, and “epithelial”.

Finally, when it comes to adverbs, there is relatively little variation in their frequency of appearance across the two fields. This suggests that adverbs play a somewhat similar, and perhaps less prominent, role in both the natural sciences and the humanities in the context of title formulation. Overall, these differences in the usage of different parts of speech mirror the unique characteristics and requirements of each academic discipline.