Navigation – Plan du site

"Activating the passive – A comparative study of the passive in scientific conference presentations and research articles

Activer le passif – Étude comparative de la diathèse passive dans des communications orales et articles scientifiques
Elizabeth Rowley-Jolivet
p. 38-52


Cet article examine l’emploi de la diathèse passive dans des communications orales de congrès scientifiques, et compare les résultats avec la fréquence constatée dans les articles correspondants publiés dans les Actes. Le corpus oral est composé de 90 communications faites en anglais par des orateurs anglophones et non-anglophones lors de congrès internationaux dans 3 domaines (géologie, médecine, physique) ; le corpus écrit comprend 18 articles publiés dans les Actes du congrès en physique. Les résultats démontrent que, parmi les facteurs qui gouvernent la fréquence d’emploi du passif, le mode (oral ou écrit) est déterminant, mais que l’origine du locuteur joue également un rôle.

Haut de page

Texte intégral


1The use of the passive in academic, and particularly in scientific, writing has been extensively studied. The reason why this particular grammatical structure has attracted so much attention is its high frequency of usage in scientific texts compared to other forms of discourse. On average, over 30 % of verbs are in the passive voice in scientific research articles (Barber 1962; Banks 1993), compared to only 2 % in spontaneous conversation (Biber et al, Longman Grammar of Spoken and Written English 1999, hereafter referred to as LGSWE), though considerable variation is found depending on the rhetorical section (Hanania & Akhtar 1985; Banks 1993), the field or speciality (Wingard 1981; Tarone et al 1998), or the genre (Master 1991). Various explanations have been proposed to account for this frequency of occurrence, ranging from logico-positivist generalisations about the ‘objectivity’ of science to considerations of thematisation, such as the importance of topic-based themes in science (Gosden 1993).

2However, scientific research communication is conducted not only in writing but also through various spoken genres. While some studies have examined aspects of didactic scientific genres such as the university seminar (Chafe 1986) and university lectures (Young 1990; Flowerdew 1994), we know remarkably little about the characteristic grammatical features of the genre that bears the closest resemblance to the research article, namely the conference presentation. It is important to investigate the differences between the research article and the conference presentation for two reasons. The first motivation concerns research into disciplinary discourse. Recent corpus-based grammars have shown (COBUILD, LGSWE) that lexico-grammatical features vary systematically according to register, with each register exhibiting preferred and dispreferred patterns of use. Across the four registers in LGSWE for example – fiction, newspaper articles, academic prose, and conversation – grammatical features are distributed in strikingly different ways. These broad register groupings provide extremely valuable yardsticks, but they need to be complemented by more narrowly focused studies of specific communicative contexts, in which register interacts with genre, creating a different configuration of grammatical characteristics. The conference presentation, for example, as an oral genre, may share some of the speech characteristics of conversation, but it is at the same time a prepared piece of research argumentation, and so may resemble the research article in other respects.

3A second reason for investigating the conference presentation concerns teaching purposes. Practically all English for Specific Purposes programmes, whatever the students’ field of study – science and engineering, economics, medicine... – now include courses on oral presentation skills, in which students receive coaching in delivering a prepared talk in their speciality, as this is recognised to be a valuable professional skill. It is clear that both the academic article and ordinary conversational practice provide unsuitable or inadequate models for this kind of discourse, and that to give students useful support, we need to know more about the linguistic features specific to this genre.

4This study therefore focuses on one particular grammatical feature, the use of the passive, in scientific conference presentations and compares the findings with its use in research articles, in order to establish what the grammatical regularities in the oral genre are. The definition of the passive used for analysis here is the syntactic one of BE + ED (following Banks 1993).

Research design

  • 1 Full details of the criteria applied to select the scientific fields, the conferences, and the pres (...)
  • 2 For technical reasons, the geology corpus comprises only 10 presentations by NS, and 20 by NNS. The (...)

5During 1993 and 1994, five international conferences in the fields of geology, medicine and physics were filmed in France. The 3 fields were chosen to represent a fairly wide spectrum of mature scientific disciplines: one life sciences discipline (medicine) and two hard sciences, the latter representing the two main loci of scientific investigation, namely fieldwork (geology) and the laboratory (physics). The specialities represented in the corpus are, respectively: i) mining geology, with a few papers from sedimentology and quaternary geology; ii) oncology; iii) electromagnetism. There is thus a bias towards the applied end of the spectrum. The language of communication was exclusively English. From the total of 185 papers recorded, 90 (30 from each field) were selected for further study and were transcribed1. The 90 papers selected are all short (15-20 mn.) presentations. Plenaries were excluded from the corpus as it was hypothesised that they may well form a different genre. In order to reflect the international origin of conference participants, native and non-native researchers (respectively, NS, NNS) were equally represented in each field2. The selected papers, recorded on video film, were then transcribed. To enable comparison between the written and spoken modes, a smaller corpus of 18 Proceedings papers (9 NS and 9 NNS), written by the same authors as those giving the oral papers, and dealing with the same subject-matter, was collected from the Book of Proceedings of the physics conference.

6The concordance software TACT 2.1 was used for certain aspects of the analysis. As the text was not tagged grammatically, the computerised analysis was complemented by a manual search. Statistical tests (ANOVA and ‘t’ tests) were run in order to establish which of the observed differences were statistically significant (at p < 0.05).

7The paper is organised as follows. The first three sections examine NS presentations only: section 1 investigates the differences between spoken and written uses of the passive by comparing 9 NS presentations and the 9 corresponding proceedings articles in physics; section 2 then extends the analysis to all 40 NS oral presentations and inquires whether there are any field-specific differences; the third section looks at the reduced passive construction (or whiz deletion), in order to highlight the relation between postmodification of the noun and the use of the reduced passive. Finally, section 4 examines the 50 oral presentations by NNS and proposes some explanations for the differences observed with NS presentations. The paper concludes with a summary of the main findings and discusses the usefulness – and the limitations – of this type of study.

1. The passive in spoken and written versions of NS conference papers

8To illustrate the considerable differences between the two genres, two short extracts are given below, one from the proceedings article, the other from the corresponding oral presentation:

  • 3 References to the data are as follows: A = article, P = presentation, followed in each case by the (...)

The polyimide plastic films manufactured by DuPont under the trademark of Kapton has been used in our stacked Blumlein pulsers as dielectric boards. It has excellent electrical properties and performs well even at high temperatures. However it is expensive and contributes a major portion of the cost of the pulser. Mylar, also manufactured by DuPont, was initially developed to serve a diverse range of industrial applications. While its electrical properties are considered to be similar to those of Kapton, its cost is substantially lower [....]. Lifetime studies of Kapton and Mylar dielectric boards were performed using a single Blumlein pulser similar to the ones developed to drive X-ray diodes. (A3-P)3

One thing we could do is go to a much cheaper dielectric, for example instead of Kapton, which is pretty expensive stuff, we could go to Mylar, which is mass-produced, readily available and really quite inexpensive. We conducted tests of both Mylar and Kapton to determine essentially how many shots we could fire into the dielectric under a particularly stressful situation, and the results were pretty dramatic (P3-P)

  • 4 See Petit (1993) for a study of some features of mathematical discourse in French and English.

9As previously mentioned, the frequency of the passive in scientific research articles has been found by various authors to be around 30 %. This figure is confirmed by that observed in the 9 proceedings articles by NS, in which the average percentage of finite clauses with a passive verb is 32.1 % (see Table 1). The figure for the 9 corresponding oral presentations, however, is over three times less (9 %). This difference is highly significant (p = 0.001). The mode – spoken or written – therefore appears to be a determining factor in the frequency of occurrence, as all other variables (author, topic, and audience, either real or invoked) are held constant. While this proportion of 9 % is higher than that of conversation, in which only 2 % of clauses use the passive voice (LGSWE), conference presentations nevertheless bear more resemblance to conversation than to academic prose in this respect. The highest percentage of passives in the oral presentations is 14 % and the spread is reasonably limited (from 4.5 % – a mathematical talk – to 14 %), whereas inter-author variation in the proceedings articles is considerably more marked (from 17 % – again a mathematical paper – to 58 %)4.

Table 1. Passives in spoken & written versions of NS conference papers

Table 1. Passives in spoken &amp; written versions of NS conference papers

10In order to understand the reasons for the marked difference in the proportion of passives in the two modes overall, all the passages containing the same information in the two versions (9 articles + 9 presentations) were compared to see what the preferred grammatical structures in each mode are. From this comparison, three points stand out.

111. In several cases, such a comparison proves to be impossible, as information given in the article is omitted in the oral presentation. The omitted information most frequently falls into two categories: i) the literature review, and ii) details of the experimental method. Both of these aspects are frequently either absent (e.g. the literature review) or minimal (e.g. experimental procedure) in oral presentations for reasons of time constraints and mental processing constraints on both speaker and audience. As previous studies have pointed out, these are the sections in the research article where the highest proportion of passives is generally found; their omission or compression in the oral presentation will therefore correspondingly reduce the frequency of passives in the latter genre.

12i) Literature review:

It has been reported that the number of spark gaps can be reduced somewhat by using a high-voltage transformer pulsed from a capacitor bank [2, 3] (A3-P)

(no information in this aspect)

13ii) Experimental details:

The thyratron housing and current returns were constructed in a way to reduce overall inductance of the switch circuit and to accommodate 5.1 cm or 2.5 cm wide Blumlein copper conductors. The ground plates from each Blumlein were smoothly twisted 90° and were connected and welded to the thyratron anode disc in a coaxial configuration. The switching assembly was pressed together by the grooved and benelex plates and installed vertically as shown in Figure 1. Each of the pulse-forming Blumleins were composed of... [18 lines of description follow] (A3-P)

and when we put the whole pulser together, well, it looks something like this, we can arrange for the Blumleins also to be wrapped around perhaps one or several times, so that we can get a longer and longer pulse, and then this is what it all looks like at the thyratron end where the switching occurs. (P3-P)

142. The passive constructions used in the research article are rendered, in the oral presentations, by active verbs, generally with a personal pronoun subject:



tests were performed

we conducted tests

it can be seen that a shock front is formed

you get a shock front forming

the signal from the loop is split into two channels and processed separately

we can put our feed-up loops and get the signal out from each of the cavities

experiments are described

what we’re talking about here is...

15There is a complementary distribution between passives and personal pronoun subjects, as Table 2 shows.

16The articles contain a high percentage of passives and a very low (3.8 %) proportion of subject pronouns with personal reference; indeed, as Table 2 shows, we is practically the only one used. In the oral presentations, in contrast, the proportion of passives is very low but over one-third of all clauses (36.4 %) have a pronoun subject with personal reference, and almost half (45.9 %) have a pronoun subject of some kind (against only 10 % in the articles). An additional point is that it does not have the same referential function in the two genres. In the great majority of cases in the articles, it (5.2 % of clausal subjects) is an empty place-holder, or anticipatory it, introducing an extraposed clause (it is desirable that, it can be deduced that). In the presentations, in contrast, it (8% of clausal subjects) generally makes anaphoric reference to a concrete object or physical parameter: it [i.e. the machine] operates, it [i.e. the voltage] goes up to 600 kV....


Table 2. Distribution of passives and personal pronouns (NS)

Table 2. Distribution of passives and personal pronouns (NS)

18The preference for the active over the passive voice can be related to the context of communication. In the conference presentation, the speaker is present in person, presenting his own results to a live audience (who will ask questions afterwards...), and the passive would be perceived as creating unwanted distance, and reducing the interpersonal dimension, between the speaker and his audience. However, the fact that the same information can be presented, when speaking, in this personal fashion, indicates that the prevalent use of the passive in the research article may be to a large extent a genre convention, as the next point illustrates.

193. The genre conventions as regards hedging in the two discourses, article and presentation, are different. In the article genre, authorial comment is expressed using passive structures such as extraposition or anticipatory it (it is believed that...; it should be noted that...), with 50 % of extraposed clauses in the passive. In the oral presentations, the speaker’s comment is congruently expressed as personal pronoun agent + active verb (we believe that...). It appears that hedged expressions in the article are often no more than conventional formulae that conform to the expected style of the research article. Indeed, their regular occurrence in this genre as ‘lexical bundles’ (LGSWE), or stable collocations, and their absence in the spoken presentations, supports this interpretation.

it is shown that
it has been found that
it can be implied that
it should be noted that
it is believed that
it is well known that

we’ve demonstrated
we find in practice
we know
note that
we believe
of course,

20It is interesting to note that while mental processes (it is assumed that, certain criteria have been deduced) represent almost one-quarter of all passives in the articles (23.4%), this figure drops to only 11% in the oral presentations, where the scientists’ decisions and hypotheses are expressed in the active form (I assume, we’ve worked out). There is also a clear contrast between the use of modal auxiliaries and the passive in the article, where 20% of all passives use a modal auxiliary (each of the sections can be modified) and their very limited use (8%) in the oral presentations.

2. Use of the passive by NS in 3 different scientific fields

21The analysis was then extended from the 9 NS presentations examined in the previous section to all 40 NS presentations in the three fields.

Table 3. Mean percentage of passives in presentations in 3 fields (NS)

Table 3. Mean percentage of passives in presentations in 3 fields (NS)

22The figures given in Table 3 demonstrate that whatever the field, the percentage of passives remains consistently low, and that overall, oral presentations use one third the number of passives compared to research articles (10.7 % against 32.1 % in the proceedings articles in physics).

  • 5 A pairwise t-test gave the following probability values: Medicine/Physics, p =0.2; Geology/Physics, (...)

23The influence of the scientific field on the use of the passive, however, appears to be slight, as the field-specific differences observed are not statistically significant5. Qualitative explanations can nevertheless be proposed for the somewhat higher figures for geology and medicine (11.8 % and 11.2 % respectively) compared to physics (9.3 %). In the former two fields, there is a higher frequency of 3 types of passive related to their objects of study and aims.

241. The ‘pseudo-passive’, as in ‘X is located in...’. In both geology and oncology, spatial location (of mineral deposits and of tumours, respectively) is an important parameter. The descriptions of the terrain, found in the Introduction section of geology presentations, contain many passives of this kind. Similarly in medicine, the tumour site is regularly commented on (29 % of tumours were localised to the abdomen…). The boundary between adjectival and verbal passives has been the subject of much linguistic discussion, but with little consensus. Indeed, Banks (1993: 16-31) argues that the traditional distinction between the two is difficult to maintain, and that an alternative analysis in terms of copula BE + Complement – meaning that all passive sentences basically express relational process – is an equally feasible explanation. The ‘pseudo-passive’ is one category where the traditional distinction seems indeed hard to maintain.

252. The ‘reciprocal passive’. This is used with verbs that express commutative relationships or, in Carnet’s words:

[des verbes] dont le procès exprime un lien entre deux éléments – que ce soit une comparaison (to compare, to equate...) ou une association (to link, to relate, to associate...) [...] [L]a particularité réside bien dans le fait que ces constructions peuvent s’inverser sans modifier le sens de l’énoncé (Carnet 1996 : 73).

26The frequent use of the reciprocal passive in geology can probably be explained by the fact that it enables the researcher to state the co-occurrence of two geological phenomena without specifying which of the two is the cause and which is the consequence. This avoids making a risky claim about the genesis of phenomena when the data are incomplete or unreliable, as is often the case in geology. There seems however to be a distinction in this field between merely spatial co-occurrence (using the verb is associated with) and a causal, though unspecified, relation (using the verbs is related to, is linked to).

273. The third type of passive, specific to medicine, is related to the deontology of the field and the respective roles of doctor and patient. The agentive role of the medical practitioner is downplayed, while patients are presented not as active individuals but as groups or populations subjected to medical procedures: patients are scheduled, randomised, stratified, made salvageable, reported, recorded or published.

3. The postmodifying reduced passive (NS)

28A prevalent feature of the grammar of scientific writing is the heavy post-modification of the noun by reduced relative clauses (-ING or -ED) as in:

A BWO consists of an electron beam guided by a strong axial magnetic field propagating through a rippled-wall slow-wave structure. (A14-P)

29This can be compared with the equivalent passage in the oral presentation:

here you have a solenoid metal magnetic field coil that’s used to guide the electron beam and here you have the corrugated slow-wave structure (P14-P).

30We are interested here in the reduced passive clauses used as noun postmodifiers. These are frequent in the articles:

An intense pulsed power source based on these novel pulsers commuted with photoconductive switches is an interesting extension of this research. (A3-P)

31This structure contributes significantly to creating a high lexical density and a low verbal density in the articles. In the oral presentations, in contrast, as is characteristic of speech in general, grammatical integration is much looser, with the information being split up or fragmented into short chunks with little noun phrase post-modification, and a correspondingly higher verbal density but lower lexical density. As can be noted in Table 1, while the 9 articles and the 9 presentations analysed here have almost the same number of words (19,475 and 19,502 respectively), there is a much lower number of clauses in the articles. The verbal densities, expressed as the number of verbs per 1000 words are, respectively, 76.5 /°° (articles) and 103/°° (presentations).

32Table 4 gives the frequency of occurrence of the reduced passive, expressed in relation to the total number of clauses, for the proceedings articles and for oral presentations in each of the 3 fields. It can be seen that overall, the articles use this structure 3 times more frequently than the oral presentations (1 reduced passive postmodifier every 9 clauses, against 1 every 24 clauses in the oral data). The highest frequency in the oral presentations is found in geology (1 reduced passive every 17 clauses), and the lowest in physics (1 every 33 clauses).

Table 4. Noun postmodification by reduced passives (NS)

Table 4. Noun postmodification by reduced passives (NS)

4. Use of the passive by non-native speakers

33A first comparison was made between the use of the passive in the proceedings articles and in the corresponding 9 oral presentations by NNS. The results are given in Table 5 (the figures already given in Table 1 for NS are reproduced for comparison).

Table 5. NS and NNS use of the passive compared in articles / presentations

Table 5. NS and NNS use of the passive compared in articles / presentations

34It can be seen that while the frequency for the two groups of speakers is almost identical in the articles, the figure for NNS in the oral presentations is appreciably higher (NNS: 14 %; NS: 9 %). While not statistically significant (p = 0.1), this difference indicates that although all speakers, whatever their nationality, use far fewer passives in the oral mode, native speakers differentiate much more radically in their choice of voice between the oral and spoken modes: their ‘switch’ from passive to active, when moving from the article to the presentation, is considerably more marked than for NNS.

35There are several possible explanations for this difference. Non-native speakers are more accustomed to reading and writing scientific English than to speaking it, and therefore remain influenced by reading and writing conventions even when speaking. Alternatively, as their confidence when speaking in English is less than that of native speakers, they are reluctant to assert themselves as speakers and to use first person pronoun subjects. Cultural factors may also play a role. As various studies in cross-cultural linguistics have highlighted, the cultural norms of academic writing differ, influencing aspects such as authorial presence and the author’s degree of assertiveness, writer- or reader-oriented text, implicit or explicit metadiscourse, etc. (see for example Mauranen 1993; Duszak 1994; Cmejrkova & Danes 1997). Two extracts are reproduced below to illustrate the differences in the mode-switching capabilities of native and non-native speakers:

NNS article
Generator energy can be stored as a magnetic field energy of current during the conduction time, and then rapidly switched into the load situated downstream of the POS. (A27 - P)

NNS presentation
... the POS which is created by a plasma injected between two high-voltage electrodes... after breaking of the switch, the current which is stored here in the form of magnetic field energy can be switched and transferred to the load (P27 - P)

36There is a marked lexico-grammatical parallelism between the spoken and written version by the non-native speaker, and a high proportion of passives used in the presentation. In the following extracts from the article and presentation by a native speaker, in contrast, out of the 4 passives in the article, only one remains in the presentation (is filled): those used for spatial description (situated, positioned) become prepositions and deictics in the presentation (behind, here, there), and the stative passive (are enclosed) becomes the interactive form of the imperative (put it all in a box).

NS article
in the case of our 700 MHz transmitter, the LCO and dipole are situated at the focus of a parabolic cylinder sub-reflector, and the LCO/dipole and sub-reflector package is positioned at the focus of a larger (3.7 m diameter) parabolic dish. (...) The modulator, L-C oscillator, dipole, and sub-reflector are enclosed in a plastic box filled with insulating gas. (A2-P)

NS presentation
It looks like this, with a parabolic sub-reflector behind it but here’s the dielectric here in the center and this is the capacitor plates here and here (...) we’re now looking down on the end of the dipole, the sub-reflector, here’s the modulator, energy storage, high-voltage transformer, put it all in a plastic box and that’s filled with sulfur hexafluoride (P2 -P)

37When all 50 oral presentations by NNS are considered, the same trend as in Table 5 can be observed (see Table 6). The differences between NS and NNS here are statistically significant in geology and physics (p = 0.06 and 0.05 respectively), but not in medicine (= 0.1).

Table 6. NS & NNS use of the passive in oral presentations in 3 fields

Table 6. NS &amp; NNS use of the passive in oral presentations in 3 fields

38Moreover, in all three fields, the oral presentations with the highest percentage of passives are without exception by non-native speakers. In some cases, the influence of the speaker’s native language (in the extract given below, French), combined with the influence of written scientific style, results in a high proportion of passives with low semantic content (characterised, represented):

We speak only about the Ofoud type which is represented by specific petrography, where basic rocks represented mostly by gabbroic rocks and anorthosite are associated with acid rocks... the Ofoud massif is characterised by a wide anorthositic country represented by massive anorthosites ... this mineralisation is characterised by a strong platinum enrichment … in blue here are represented the different massifs... (P20-G)

39In a few other NNS presentations, the very high proportion of passives ( 30 % of all clauses) is due to the fact that the speaker reads a fully-written out text, or recites a text that has been learnt by heart. The following extract shows that reciting a text results in a stilted style, lacking in interactivity, with a high information density using non-congruent structures which take more time to process for the listener:

However, it can be noticed that sulphides extracted from the laminated dolomite member, that’s to say from lower deposits, provide less radiogenic values than the sulphides extracted from higher deposits hosted by the Ceroïde limestone member. At first sight it can be inferred from these data that these deposits are syngenetic deposits of Cambrian age... (P30-G)

Summary and Comment

40This study has investigated the use of the passive in scientific conference presentations in order to see whether its occurrence in spoken research science resembles that found in scientific writing. The results show clearly that for native speakers, the passive is not a salient or characteristic feature of spoken science, representing on average only 10.7 % of verb choices. The influence of the mode – spoken or written – therefore appears to be a highly significant variable governing the choice between active and passive in scientific research.

41We have also observed a complementary distribution between passives and personal pronoun subjects in the articles and presentations, and have related this to the enunciative context and the different conventions of the two genres. Field-specific differences appear, however, to have little impact on the use of the passive in the three fields studied here, although some qualitative differences are observed.

42Lastly, the data show a higher, and statistically significant, proportion of passives, irrespective of the field, in NNS conference presentations compared to those by native speakers, and several explanations have been proposed to account for this difference. It is hoped that the size of the corpus used for this study – 90 presentations, 40 by NS, 50 by NNS, drawn from 3 different fields – forms a sufficiently large and varied database for the conclusions summarised above to have general validity.

43Quantitative data of the kind presented here, however, are just the first step in analysing the characteristic features of spoken scientific discourse. They provide a useful benchmark, and demonstrate that the grammatical regularities of written and spoken science are radically different – the research article is clearly an unsuitable model for giving an oral presentation – but they need to be complemented by further discourse-based studies.

44Some aspects of the passive not considered here but which are worthy of further study in academic speech are: What functions does the passive fulfil in the oral presentations? Secondly, given the low proportion of passives, what does this imply about the speaker-audience relationship or the interpersonal dimension in research presentations? (see Ventola et al eds. forthcoming, for several articles on this aspect). The passive is one of a number of specialised syntactical structures which are used to manipulate theme choice in the clause, and plays a role both in information salience and in textual cohesion: in view of the relative rarity of the passive in the oral presentations, what strategies and structures are used instead to ensure salience and cohesion? (Carter-Thomas & Rowley-Jolivet 2001).

45Despite the limitations of a study of this kind, it throws an interesting light on a vexed epistemological question: much has been made of the ‘objectivity’ or ‘impersonality’ of science, and the prevalence of the passive in written science has been frequently adduced to support this point of view. The study of spoken scientific discourse, however, indicates that the passive is to a large extent a mode-governed feature. Whether this should be interpreted as showing that the use of the passive is merely a stylistic convention in the research article, or whether the observed differences in use are related to the different status of the knowledge claims made in the research article and in the conference presentation, remains for the moment an open question (but see Rowley-Jolivet, forthcoming, in Ventola et al, for some discussion of this).

46An additional interest of this study concerns implications for the teaching of oral presentation skills to non-native speakers. As the comparison between the use of the passive by native and non-native speakers shows, non-native speakers retain, even when speaking, many of the syntactical features of scientific writing, resulting in a less interactive style of delivery: their awareness of what is involved in mode-switching, and of the processing constraints on the audience in the case of real-time delivery needs to be developed using comparative data of the type presented here.

47Banks, David. 1993. Writ in Water. Aspects of the Scientific Journal Article. E.R.L.A. Université de Bretagne Occidentale.

48Barber, C. L. 1962. “Some measurable characteristics of modern scientific prose”. Reprinted in Swales, J. (Ed.). 1985. Episodes in ESP. Pergamon Press.

49Biber, D., S. Johansson, G. Leech, S. Conrad, E. Finegan. 1999. Longman Grammar of Spoken and Written English. Harlow: Pearson Education Ltd.

50Carnet, D. 1996. Opérations de détermination caractéristiques de l’anglais médical. Thèse de doctorat, Université de Bourgogne.

51Carter-Thomas, S. & E. Rowley-Jolivet. 2001. “Syntactical differences in oral and written scientific discourse: The mediating power of information structure”. Communication au XXIIe Colloque du GERAS, Bordeaux.

52Chafe, W. 1986. “Academic speaking”. Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society. Berkeley, California: University of California: 26-40.

53Cmjrkova, S. & D. Frantisek. 1997. “Academic writing and cultural identity: The case of Czech academic writing”. In Duszak, A. (ed.). Culture and Styles of Academic Discourse. Berlin: Mouton de Gruyter: 41-61.

54Duszak, A. 1994. “Academic discourse and intellectual styles”. Journal of Pragmatics 21: 291-313.

55Flowerdew, J. (Ed.). 1994. Academic Listening. Cambridge: Cambridge UP.

56Gosden, H. 1993. “Discourse functions of subject in scientific research articles”. Applied Linguistics, 14/1: 56-75.

57Hanania, E. & K. Akhtar. 1985. “Verb form and rhetorical function in science writing: A study of MS theses in biology, chemistry and physics”. The ESP Journal 4: 49-58.

58Master, P. 1991. “Active verbs with inanimate subjects in scientific prose”. English for Specific Purposes 10: 15-33.

59Mauranen, A. 1993. “Contrastive ESP rhetoric: Metatext in Finnish-English economics texts”. English for Specific Purposes 12: 3-22.

60Petit, M. 1993. Analyse contrastive de la relation discursive et interénonciative dans un corpus d’ouvrages mathématiques en français et en anglais. Thèse de doctorat, Université de Paris 7.

61Rowley-Jolivet, E. 1998. La Communication scientifique orale. Étude des caractéristiques linguistiques et discursives d’un genre. Thèse de doctorat. Université de Bordeaux 2.

62Rowley-Jolivet, E. Forthcoming. “Science in the making. Scientific conference presentations and the construction of facts”. In Ventola, E. et al (eds.) Conference Language in Europe.

63Tarone, E., S. Dwyer, S. Gillette, V. Icke. 1998. “On the use of the passive and active voice in astrophysics journal papers: With extensions to other languages and other fields”. English for Specific Purposes 17/1: 113-132.

64Ventola, E., S. Thompson, C. Shalom (Eds.). Forthcoming. Conference Language in Europe. Peter Lang.

65Wingard, P. 1981. “Some verb forms and functions in six medical texts”. In Selinker, L., E. Tarone & V. Hanzeli (Eds.). English for Academic and Technical Purposes: Studies in Honor of Louis Trimble. Rowley, MA: Newbury House: 53-64.

66Young, L. 1990. Language as Behaviour, Language as Code. Amsterdam: John Benjamins.

Haut de page


1 Full details of the criteria applied to select the scientific fields, the conferences, and the presentations, can be found in (Rowley-Jolivet 1998: Chapter 4).

2 For technical reasons, the geology corpus comprises only 10 presentations by NS, and 20 by NNS. The medicine and physics corpora each have 15 NS and 15 NNS presentations.

3 References to the data are as follows: A = article, P = presentation, followed in each case by the number of the article or presentation in the corpus and then a letter indicating the field (G = geology, M = medicine, P = physics).

4 See Petit (1993) for a study of some features of mathematical discourse in French and English.

5 A pairwise t-test gave the following probability values: Medicine/Physics, p =0.2; Geology/Physics, p = 0.2; Geology/Medicine, p = 0.7.

Haut de page

Table des illustrations

Titre Table 1. Passives in spoken & written versions of NS conference papers
Fichier image/png, 29k
Titre Table 2. Distribution of passives and personal pronouns (NS)
Fichier image/png, 136k
Titre Table 3. Mean percentage of passives in presentations in 3 fields (NS)
Fichier image/png, 27k
Titre Table 4. Noun postmodification by reduced passives (NS)
Fichier image/png, 60k
Titre Table 5. NS and NNS use of the passive compared in articles / presentations
Fichier image/png, 14k
Titre Table 6. NS & NNS use of the passive in oral presentations in 3 fields
Fichier image/png, 30k
Haut de page

Pour citer cet article

Référence papier

Elizabeth Rowley-Jolivet, « "Activating the passive – A comparative study of the passive in scientific conference presentations and research articles », Recherche et pratiques pédagogiques en langues de spécialité, Vol. XX N° 4 | 2001, 38-52.

Référence électronique

Elizabeth Rowley-Jolivet, « "Activating the passive – A comparative study of the passive in scientific conference presentations and research articles », Recherche et pratiques pédagogiques en langues de spécialité [En ligne], Vol. XX N° 4 | 2001, mis en ligne le 27 août 2014, consulté le 23 janvier 2020. URL : ; DOI : 10.4000/apliut.4750

Haut de page


Elizabeth Rowley-Jolivet

Maître de Conférences à l’École Supérieure des Procédés Electroniques et Optiques (ESPEO), Université d’Orléans.

Articles du même auteur

Haut de page

Droits d’auteur

Association des Professeurs de Langues des Instituts Universitaires de Technologie

Haut de page