1. Introduction and historical background
The first ever periodical of an academic nature was the Journal des Sçavans, which first appeared on 5 January 1665 in Paris. It was edited by Denis de Sallo at the instigation of Colbert, Louis XIV’s Contrôleur général des finances (the most powerful position after that of the king himself), with the objective of controlling new knowledge, which he perceived as being a potential danger to the State. It covered the whole range of academic disciplines of the period, and was made up mainly of book reviews (Banks 2015, 2017a, Morgan 1928). Only two months later, on 6 March 1665, the first issue of the Philosophical Transactions appeared in London. It was edited by Henry Oldenburg, and although it had the imprimatur of the Royal Society, it remained the personal property of Oldenburg, who retained full editorial and financial responsibility for it. In addition to being one of the secretaries of the Royal Society, he was the centre of a network of scientific correspondence, and he hoped that the creation of a bulletin of scientific information would be a source of extra income. Its contents was restricted to science and technology and depended largely on the contents of his postbag (Banks 2017a, Hall 2002, Bluhm 1960). The Royal Society had been created in 1660, but, despite having royal patronage, did not receive any official funding and depended on the subscription fees of its members for its income. This meant that many members were elected because of their ability to pay, rather than for any scientific ability (Lyons 1944, Bluhm 1958, Hunter 1982). The Académie Royale des Sciences was created in 1666, but unlike its English counterpart was created and maintained by the state which also provided premises in which to carry out their activities, and its members were paid a retainer by the crown (Hahn 1971, Hirschfield 1981). Although the Académie did publish its Mémoires, this was done only in luxurious limited editions which were considered the personal property of the King, who frequently used them as gifts for illustrious visitors. The fact that this was seen to be part of the glory and aura of the French monarchy meant that it was felt that wider distribution was unnecessary (Licoppe 1994). This situation lasted until the end of the century which meant that during this period the Journal des Sçavans and the Philosophical Transactions remained the two major scientific periodicals publishing in vernacular languages.
It was only after the reorganization of the Académie in 1699 that this situation changed radically. It was then decided that the Mémoires of the Académie should be published on a much more regular basis, and the volume for 1699 actually appeared in print in 1702 (Hahn 1971, McClellan 2001, Banks 2017b). The Philosophical Transactions remained the personal property of its successive editors until 1752. However, from 1730 onwards, it did not make a profit and the Royal Society eventually took over responsibility for the publication in 1752. The Royal Society continued to run it at a loss because it enhanced its reputation and was important in its reciprocal relations with groups and individuals abroad. From that date the Philosophical Transactions was run by an editorial committee, who decided what to publish; all papers published after this date had been previously read at a Royal Society meeting (Fyfe 2015, Fyfe et al. 2015, Fyfe & Moxham 2016).
In the 1790s France underwent the trauma of the French Revolution. From the end of 1792 scientific activity was greatly reduced. There was great suspicion of intellectuals, and, despite the efforts of Lavoisier, all académies, including the Académie Royale des Sciences were abolished in 1793. Lavoisier himself, despite being at the forefront of scientific activity and with an international reputation, was arrested in November of that year, and executed a few months later, a victim of Marat’s vindictiveness as much as any misdemeanours related to his tax-gathering activities. In 1779 Lavoisier had been instrumental in discrediting Marat’s scientific pretentions (Bell 2005, Hahn 1971). Marat believed that he was the greatest scientist of his age, and that he was even greater than Newton.
Having made his office in the rue de Bourgogne into a laboratory, he carried out numerous experiments on light, fire and electricity, and soon convinced himself that he was the man promised to Science to revolutionize the whole of physics and relegate Newton to the domain of illusion. (Fayet 1960: 29)1
The backlog of volumes of the Mémoires continued to appear: the volume for 1789 was printed in 1793, but not distributed until 1796, with the word “Royale” still present in its running head since it would have been too expensive to alter the presses; the volume for 1790 appeared in 1797 (Hahn 1971).
The abolition of the académies created a vacuum, which was filled in 1795 with the creation of the Institut National des Sciences et des Arts. This had three classes, the first of which dealt with physical and mathematical sciences, thus providing a replacement for the defunct Académie Royale des Sciences, and they began publishing Mémoires from 1798. In 1816, the Académie regained its independence, and its name (Académie des Sciences), within the Institut de France. By this time France had undergone not only the Revolution, but also the rule of Napoleon, both during the Consulat, 1799-1804, and the Empire, 1804-1814, and the Restoration of the Monarchy under Louis XVIII in 1814. Nevertheless, according to Brockliss the Revolutionary and Napoleonic eras were accompanied by an “explosion of scientific creativity” (1992, 79).
The Journal des Sçavans also fell foul of the French Revolution. It disappeared in 1792. There was an unsuccessful attempt to revive it in 1796. It appeared regularly for a few months, but then disappeared again. It was only in 1816, after the Restoration of the monarchy, that the Journal des Sçavans was finally re-established on a more solid and lasting basis (Cocheris 1860, Paris 1903).
England was not, of course, directly affected by the French Revolution. The Royal Society was dominated during this period by Joseph Banks, its longest serving President, who was elected in 1778, at the relatively early age of 35, and remained in control till his death in 1820 (Lyons 1944, Cameron 1952, O’Brian 1987). He had come to fame as the botanist who had accompanied Captain Cook on his first circumnavigation of the world (Beaglehole 1962a, b, Banks 1996). He is frequently seen as being an autocratic ruler who directed the Royal Society with an iron hand, but while Valle (1999) finds that botany and zoology were dominant in the period 1760 to 1820, Banks does not seem to have given special favour to his own subject, biology, although he did have a preference for the applied sciences (Gascoigne 1994). England and France were intermittently at war from the middle of the eighteenth century until the Battle of Waterloo in 1815; this included the Seven Years War (1756-1763), the American War of Independence (1777-1783) in which the French actively supported the Americans, and the period of the French Revolution and the Napoleonic Wars (1793-1815). In spite of this the Royal Society and the Académie remained in contact. Banks himself claimed that science was above politics, and is credited with the quote that the “sciences were never at war” (Crosland 2005). He became a corresponding member of the Académie in 1772, and in 1802, he was elected to the Institut National (Gascoigne 1994, 1998). Throughout this period the situation of the Philosophical Transactions remained stable, and while the Royal Society went through an unstable period following Banks’ death, this does not seem to have affected the Philosophical Transactions itself (Atkinson 1999).
It has been shown that text is very closely related to, and to some extent caused by, the context in which it is produced (Halliday 1978, Banks 2005, 2017a, c). In Banks 2017a, I showed that there were differences in the linguistic features used in the writing of the early issues (1665-1700) of the Journal des Sçavans and the Philosophical Transactions. These differences can be traced to the different editorial decisions made by de Sallo and Oldenburg in relation to the genre to be used (mainly book reviews or extracts from letters), and the field to be covered (all known knowledge or only science and technology). These decisions themselves can be seen as a function of the very different historical situations of France and England in the late seventeenth century. The object of this paper is to attempt to see to what extent (if any) scientific writing was affected by the turbulence of this period: the French Revolution in France, and the wars between England and France.
2. A corpus
There seem to be three main contenders for consideration in this study: the Journal des Sçavans, the Mémoires of the French Académie, and the Philosophical Transactions. This does not mean that no other periodicals were created; on the contrary there were vast numbers of publications devoted to scientific subjects. However, many of these were short-lived. Kronick (1962) indicates that of almost 500 scientific periodicals in the period 1670 to 1700, only 31% lasted more than five years, and only 18% more than ten. In contrast, the Journal des Sçavans and the Philosophical Transactions still exist, and the Mémoires continued to be published until 1936.
I have selected the years 1785 and 1835 for analysis. This is motivated by the fact that this gives us a round 50-year period, 1785 precedes the French Revolution, and 1835 is 20 years after the Battle of Waterloo; there is also the practical reason that the issues of the periodicals for these years are easily accessible.2
In its issues for 1785 the Philosophical Transactions printed 25 items. Of these, 20% were in the field of astronomy, and a further 20% in the field of physics, with 12% of the items being devoted to biology and 12% to chemistry. In the same year, the volume of the Mémoires contained 49 items, 28% of which concerned chemistry, 22% astronomy, 12% physics, and 10% biology. Hence, for that year, 64% of the items printed in the Philosophical Transactions and 72% of those in the Mémoires were devoted to the four main fields of astronomy, chemistry, physics, and biology. With the possible exception of two papers devoted to philosophical subjects in the Mémoires, all of the items can be considered as being within the general field of science and technology. The full details are given in Table 1.
Table 1. Subjects in Philosophical Transactions and Mémoires, 1785.
| Phil Trans | Mémoires | |||
| 1785 | 1785 | |||
| Astronomy | 5 | 20% | 11 | 22% |
| Archaeology | - | - | - | - |
| Biology | 3 | 12% | 5 | 10% |
| Botany | 2 | 8% | 4 | 8% |
| Chemistry | 3 | 12% | 14 | 28% |
| Geography | 1 | 4% | - | - |
| Geology | 2 | 8% | 1 | 2% |
| Mathematics | 1 | 4% | 1 | 2% |
| Medicine | - | - | 2 | 4% |
| Meteorology | 1 | 4% | 1 | 2% |
| Physics | 5 | 20% | 6 | 12% |
| Philosophy | - | - | 2 | 4% |
| Technology | 2 | 8% | 2 | 4% |
| 25 | 49 | |||
The Journal des Sçavans, on the other hand, published 148 items during this period, of which 16% were of a general nature or covered more than one subject area. History accounts for 11% of the items, geography for 10% and meteorology for 9%. Indeed, all scientific subjects taken together account for only 37% of the total, the most important being geography, 10%, and meteorology, 9%; these two subjects are quite marginal in the Philosophical Transactions and the Mémoires. I shall take it then that the Journal des Sçavans is not directly comparable to the other two publications and it will not be considered further here. The full details are given in Table 2, where scientific subjects (i.e. those that are represented in the Philosophical Transactions and the Mémoires) are indicated by an asterisk.
Table 2. Subjects in Journal des Sçavans, 1785.
| JdS | ||
| 1785 | ||
| Agriculture | 2 | 1% |
| Anthropology | 3 | 2% |
| Archaeology* | 1 | 1% |
| Astronomy* | 1 | 1% |
| Bibliography | 1 | 1% |
| Biology* | 1 | 1% |
| Botany* | 2 | 1% |
| Chemistry* | 3 | 2% |
| Classics | 4 | 3% |
| Edition | 2 | 1% |
| Education | 1 | 1% |
| General | 23 | 16% |
| Geography* | 15 | 10% |
| History | 17 | 11% |
| Language | 1 | 1% |
| Law | 9 | 6% |
| Mathematics* | 3 | 2% |
| Medicine* | 9 | 6% |
| Meteorology* | 13 | 9% |
| Music | 2 | 1% |
| Numismatics | 1 | 1% |
| Obituary | 3 | 2% |
| Physics* | 2 | 1% |
| Technology* | 5 | 3% |
| Theology | 8 | 5% |
| Technology* | 5 | 3% |
| Theology | 8 | 5% |
| Total | 148 | |
In its issues for 1835, the Philosophical Transactions printed 21 items, of which 32% were devoted to physics, and 18% each to biology and geography. The volume of the Mémoires for that year has only 12 items, 25% of which deal with biology, 25% with chemistry, and 17% with botany. Interest in all the main areas seems to have been maintained, but with changing focus. Chemistry has about the same degree of interest; interest in biology has increased, while that for astronomy seems to have decreased; interest in physics and geography has increased for the writers in the English publication but not in the French, while the opposite is true of botany. The full details are given in Table 3.
Table 3. Subjects in Philosophical Transactions and Mémoires, 1835.
| Phil Trans | Mémoires | |||
| 1835 | 1835 | |||
| Astronomy | 1 | 5% | 1 | 8% |
| Archaeology | - | - | 1 | 8% |
| Biology | 4 | 18% | 3 | 25% |
| Botany | - | - | 2 | 17% |
| Chemistry | 1 | 5% | 3 | 25% |
| Geography | 4 | 18% | - | - |
| Geology | 1 | 5% | 1 | 8% |
| Mathematics | 1 | 5% | - | - |
| Medicine | 2 | 9% | - | - |
| Meteorology | 1 | 5% | - | - |
| Physics | 7 | 32% | - | - |
| Technology | - | - | 1 | 8% |
| Total | 21 | 12 | ||
For the purposes of analysis a number of articles were selected for each of the two periodicals at each of the two time points. An attempt was made to include the scientific fields which were most commonly represented in the periodicals in question, but articles of more than 12 pages were excluded. This gave, for the Philosophical Transactions 1785, three articles totalling 27 pages, for the Philosophical Transactions 1835, five articles totalling 26 pages, for the Mémoires 1785, six articles totalling 24 pages, and for the Mémoires 1835 three articles totalling 28 pages. The only one of these which causes any real difficulty is that for the Mémoires 1835. The articles in this sample are considerably longer than in the other sub-corpora, which meant that many were excluded on grounds of length. Of the three selected, two are by the same author (Becquerel). While this is not ideal, the results do not seem to indicate any particular skewing due to this feature. Details of the corpus are given in Appendix 1.
3. A linguistic framework
I intend to look at two features in these texts, the nature of themes, and process types. The analysis is based on the principles of Systemic Functional Linguistics (Halliday 2014, Banks 2005, 2017c). Systemic Functional Linguistics sees the clause as embodying three types of meaning, or metafunctions: ideational, which deals with processes, the participants involved in them, and their associated circumstances; interpersonal, dealing with the relationships established by the speaker with his addressees, and with his own message; and the textual, which deals with the organization of the message.
Theme is a component of the clause and is defined as the speaker’s starting point. It constitutes part of the thematic structure and falls within the domain of the textual metafunction. In both French and English it occurs in initial position. Each clause has a topical theme, which is the first major component of the clause (subject, circumstantial adjunct, complement or predicator). This may be preceded by an interpersonal and/or textual theme, though neither of these is necessarily present. In this study the term “theme” without further specification should be taken to mean topical theme. Hence in (1) For some years past functions as circumstantial adjunct and is the theme of the clause. If Dear Sir is considered to be part of this clause it functions as an interpersonal theme.
(1) DEAR SIR, For some years past I have been employed in verifying all the stars suspected to be variable, in order that hereafter we may know with certainty what to depend upon. (Pigott, PT 1785)3
In (2), Les veines principales de la grenouille is the subject and functions as theme.
(2) Les veines principales de la grenouille ont donc un mouvement ou battement propre. (Flourens, Mémoires 1835)4
Transitivity concerns processes, participants and circumstances, and is the major component of the ideational metafunction. For process types, I shall consider the processes encoded by the finite verbs. To the extent that the interpretation of process types can vary from grammatical to conceptual (O’Donnell et al. 2008), my view is towards the conceptual pole of this cline. I use a system which has five process types (Banks 2005, 2017c). Material processes are actions or events of a physical nature.
(3) Such observations were accordingly made at all the Preventive Service stations on the coasts of England, Ireland, and Scotland, from June 7 to June 22 inclusive, and the registers of the observations were sent to the Admiralty, where they now are. (Whewell, PT 1835)
(4) Dans les poissons, la circulation ne s’opère pas de la même manière que dans les animaux qui respirent de l’air. (Broussonet, Mémoires 1785)5
Mental processes are activities of a cerebral type; these can be cognitive, affective, or processes of perception.
(5) The refrangibility was seen to vary considerably and irregularly for each ray and each medium; and when FRAUNHOFFER had assigned serieses of numbers as the accurate expressions of the varying refractive powers throughout the several spectra, the apparent absence of any law connecting these numbers was only rendered more palpable. (Powell, PT 1835)
(6) Je présume que telle a été la marche du tonnerre, d’après l’inspection du local & l’examen des effets. (Tessier, Mémoires 1785)6
Relational processes link an entity to another entity or one of its own characteristics. These can be attributive, identifying or possessive.
(7) A very brief examination of its optical was sufficient to indicate its more obvious peculiarities, and a short notice was published at the time. (Brewster, PT 1835)
(8) Ces deux actions sont d’autant plus grandes que l’on a diminué davantage la force d’agrégation. (Bequerel, Mémoires 1835)7
Verbal processes are processes of communication.
(9) The extension of such results as have been stated in the present paper to other coasts, and the discovery of other similar laws, cannot but be looked upon as a valuable and interesting addition to our knowledge. (Whewell, PT 1835)
(10) Un caractère que l’on distingue plus fortement encore est celui que je vais indiquer. (Flourens, Mémoires 1835)8
Existential processes simply state the existence of an entity.
(11) … and therefore there is no reason to think, that the precipitate, which our salt occasioned with a solution of silver, proceeded from any other cause than its being phlogisticated. (Cavendish, PT 1785)
(12) Dans la perche il n’y a qu’un seul ovaire, l’artère spermatique est unique … (Broussonet, Mémoires 1785)9
I do not use the category of behavioural process proposed by many working within the Systemic Functional framework (Banks 2016).
4. Theme
As stated above, Systemic Functional Linguistics analyses the thematic structure of a clause in terms of a theme and a rheme. The theme is the speaker’s starting point and in languages like French and English is placed in initial position. Thematic structure has been seen as the driving force of scientific writing (Banks 2008a,b, Halliday & Martin 1993). Halliday (1988) has noted the importance of thematic structure in the construction of the argument of a scientific text, whereby the rheme of one clause is taken up as the theme of a subsequent clause. This is particularly significant when the theme nominalizes a segment which appeared in a non-nominalized form in the preceding rheme. I shall look first at the grammatical functions of the themes in the corpus, and then at the semantic categories that they represent.
The unmarked grammatical function of a theme is subject, and this is borne out in the corpus. The results for the Mémoires are given in Table 4.
Table 4. Grammatical function of themes in Mémoires.
| 1785 | 1835 | |||
| S | 243 | 76% | 135 | 70% |
| Adj | 62 | 19% | 47 | 24% |
| P | 2 | 1% | 4 | 2% |
| C | - | - | - | - |
| Extra | 10 | 3% | 6 | 3% |
| Cleft | 2 | 1% | 1 | 1% |
| Total | 319 | 193 | ||
As can be seen, the vast majority of themes function as subject, 76% in 1785 and 70% in 1835. The only other function of any note is that of adjunct, which accounts for 19% of themes in 1785 and 24% in 1835. The other categories never account for more than 3%, and there are no examples of complement themes in the Mémoires corpus. There is little change over time other than a slight reduction in the percentage of subject themes balanced by a similar rise in adjunct themes.
The corresponding results for the Philosophical Transactions are given in Table 5.
Table 5. Grammatical function of themes in Philosophical Transactions10.
| 1785 | 1835 | |||
| S | 122 | 54% | 290 | 66% |
| Adj | 88 | 39% | 127 | 29% |
| P | 1 | * | 1 | * |
| C | 5 | 2% | 2 | * |
| Extra | 11 | 5% | 17 | 4% |
| Cleft | - | - | - | - |
| Total | 227 | 437 | ||
In the Philosophical Transactions the subject themes are still by far the largest category, but rather less so than in the Memoires, accounting for 54% in 1785 and 66% in 1835. Once again, the only other function of note is that of adjunct, which accounts for 39% of the themes in 1785 and 29% in 1835. The other categories never account for more than 5%, and there are no examples of cleft structures providing themes in the Philosophical Transactions corpus. Here, there is an increase in the percentage of subject themes at the expense of adjunct themes. Moreover the increase (12 percentage points) is rather greater than the decrease in the Mémoires (6 percentage points). So, the change in the Philosophical Transactions is greater than that in the Mémoires, although the Philosophical Transactions might be seen as moving towards the profile provided by the Mémoires.
The semantic categories of themes which I shall use are based on those developed in Banks 2008a. At the risk of being fastidious, I shall give a brief outline of the categories, with examples from the corpus. They are as follows:
Obj. The object of study or experimentation.
(13) The non-conducting power of a perfect vacuum is a fact in electricity which has been much controversed among philosophers. (Morgan, PT 1785)
(14) Le pouls veineux dont il s’agit ici est un phénomène d’un tout autre ordre. (Flourens, Mémoires 1835)11
Exp. The experiment or the experimental process.
(15) These experiments, however, belong to another subject, and may possibly be communicated at some future time. (Morgan, PT 1785)
(16) Cette opération, dans laquelle j’ai employé quatre onces d’acide marin concentré & une once de manganèse, étant finie, je n’ai point trouvé que l’esprit-de-vin eût l’odeur d’acide marin déphlogistiqué. (Berthollet, Mémoires 1785)12
Equip. The equipment used.
(17) Diaphragms were attempted; but besides other difficulties, they did not efface stars of the first magnitude. (Pigott, PT 1785)
(18) La lunette de cet instrument est foible, & le micromètre dont elle est garnie est de beaucoup inférieur à ceux qu’on a construits depuis. (Jeaurat, Mémoires 1785)13
Obs. The process of observation.
(19) Such observations were accordingly made at all the Preventive Service stations on the coasts of England, Ireland, and Scotland, from June 7 to June 22 inclusive, and the registers of the observations were sent to the Admiralty, where they now are. (Whewell, PT 1835)
(20) L’OBSERVATION du passage de la Lune, par la constellation des Pléïades, est une de celles qui peuvent être les plus propres à fixer, pour nos tables de la Lune, nombre de secondes d’erreur, dont elles sont encore susceptibles. (Jeaurat, Mémoires 1785)14
Auth. The author of the text in question or a group of which he is a member.
(21) I accordingly made some experiments to determine what degree of purity the air should be of, in order to be diminished most readily, and to the greatest degree. (Cavendish, PT 1785)
(22) Nous avons fait saigner du bras les deux ‘premiers’, & le troisième du pied, environ six heures après l’évènement. (Tessier, Mémoires 1785)15
Oth. Humans other than the author.
(23): Sir EVERARD HOME has given a very good representation of them, at this stage of their progress, in his Lectures on Comparative Anatomy, from the elegant pencil of Mr. BAUER. (Thompson, PT 1835)
(24) M. PRIESTLEY a retiré du gaz inflammable de l’esprit-de-vin, par le moyen de l’étincelle électrique. (Berthollet, Mémoires 1785)16
Inter. Documents other than the present text.
(25) For instance, in a paper published in the Philosophical Transactions for 1819, it is taken for granted, that when the two tides meet which come up the British Channel and down the German Ocean there must be a visible and marked conflict of opposite currents of water. (Whewell, PT 1835)
(26) De plus, les tables du Soleil de l’Abbé de la Caille, donnent 3,5" de plus que celle de Mayer. (Jeaurat, Mémoires 1785)17
Meta. Reference to other parts of the same text.
(27) My next illustration will show that a similar fact has been observed among the bivalves. (Gray, PT 1835)
(28) Les figures (1), (2), (3) montrent ce qui se passe dans l’arrangement des atomes pendant les premiers moments de la cémentation. (Bequerel, Mémoires 1835)18
Exist. Existential themes.
(29) There are some circumstances which we can readily imagine may produce such an effect, though awe should probably not succeed in guessing what the effect would be. (Whewell, PT 1835)
(30) … il n’y en a eu que deux qui soient restés sur leurs pieds. (Tessier, Mémoires 1785)19
Ment. Mental processes or elements of argumentation.
(31) But this conclusion has not been universally admitted, and the following experiments were made with the view of determining its truth or fallacy. (Morgan, PY 1785)
(32) En continuant le même raisonnement, on arrivera à l’arrangement de la figure 3, et ainsi de suite. (Becquerel, Mémoires 1835)20
Time. Temporal expressions.
(33) In 1778 I had small pieces of fine glass stained with different shades, which being applied to the eye end of a telescope, I could easily find what degree of shade was requisite to efface stars of different brightness. (Pigott, PT 1785)
(34) Le 13, à 11 heures 52 minutes du soir, des nuages s’élevèrent du sud, couvrirent la Lune, de manière que toutes les étoiles qui étoient autour d’elle disparurent. (Messier, Mémoires 1785)21
Sit. Geographical or spatial location.
(35) From the Scilly Islands to Portland Bill, most of the stations exhibit this inequality operating upon the greater part of the tides. (Whewell, PT 1835)
(36) Vers le nord, le tonnerre a cassé des tuiles, depuis le faîte jusqu’à l’égout, & a réduit en copeaux un chevron dans toute sa longueur. (Tessier, Mémoires 1785)22
Math. Mathematical expressions.
(37) And the variation in the value of this ratio, and consequently in that of μ, for a given variation in λ, is greater when the arc is greater, this, when λ is less. (Powell, PT 1835)
(38) A ce résultat j’ajoute, pour dernière remarque, qu’à cette époque, 13 décembre 1785, & pour 9h 59' 7", temps vrai, les tables du Soleil de Mayer & de l’Abbé de la Caille différèrent entr’elles seulement de 3,5". (Jeaurat, Mémoires 1785)23
Table 6 gives the distribution of these categories for the Mémoires.
Table 6. Semantic categories of themes in the Mémoires.
| 1785 | 1835 | |||
| Obj | 191 | 60% | 119 | 62% |
| Exp | 9 | 3% | 8 | 4% |
| Equip | 4 | 1% | - | - |
| Obs | 5 | 2% | 3 | 2% |
| Auth | 59 | 18% | 31 | 16% |
| Oth | 17 | 5% | 7 | 4% |
| Inter | 1 | * | 4 | 2% |
| Meta | - | - | 1 | 1% |
| Exist | 6 | 2% | - | - |
| Ment | 10 | 3% | 19 | 10% |
| Time | 14 | 4% | - | - |
| Sit | 2 | 1% | 1 | 1% |
| Math | 1 | * | - | - |
| Total | 319 | 193 | ||
There is a striking similarity between the results for the two dates. Hence over this 50-year period there seems to have been virtually no change in the way these writers choose their themes. The only relatively small difference is the 7 percentage point rise for the Ment category. Otherwise none of the categories change by more than 4 percentage points, and most of them by only one or two. By far the largest category is Obj, showing the importance of the object of study for these writers, who place it in thematic position in almost two out of three clauses. The second most frequent category is that of Auth, making the author himself the second most important type of theme, and accounting for almost one in five of the clauses. No other category accounts for as many as 10% of the themes, except Ment in 1835.
Table 7 gives the corresponding distribution for the Philosophical Transactions.
Table 7. Semantic categories of themes in the Philosophical Transactions.
| 1785 | 1835 | |||
| Obj | 101 | 44% | 257 | 59% |
| Exp | 26 | 11% | 2 | * |
| Equip | 15 | 7% | - | - |
| Obs | 5 | 2% | 13 | 3% |
| Auth | 31 | 14% | 31 | 7% |
| Oth | 4 | 2% | 33 | 8% |
| Inter | 5 | 2% | 12 | 3% |
| Meta | - | - | 3 | 1% |
| Exist | 4 | 2% | 5 | 1% |
| Ment | 29 | 13% | 58 | 13% |
| Time | 7 | 3% | 6 | 1% |
| Sit | - | - | 8 | 2% |
| Math | - | - | 9 | 2% |
| Total | 227 | 437 | ||
Here, there seems to be rather more change over time than was the case in the Mémoires. The Obj category is again by far the largest, though not to the same extent as in the Mémoires. Moreover, the frequency of the category increases from 44% to 59% (15 percentage points) over the period 1785 to 1835. However, this apparently large increase must be tempered to some extent. If one takes as a group Obj + Exp + Equip + Obs, which might be thought of as those categories directly connected to the study itself, then we have 65% for 1785, compared with 62% for 1835. The corresponding figures for the Mémoires are 66% in 1785, and 67% in 1835. Hence although the Obj category is less frequent in the Philosophical Transactions and increases over time, the study group as a whole changes little, and has similar frequencies in the two periodicals. The Auth category in the Philosophical Transactions falls from 14% to 7% (7 percentage points), while the Oth category increases by almost the same amount from 2% to 8% (6 percentage points). These two categories hardly changed at all over time in the Mémoires. It is possible that what was to come to be seen as the impersonal nature of scientific discourse was beginning to be felt in the early nineteenth century, leading to a reduction in the use of Auth themes in English. Similarly it is possible that the need to reference the work of others was beginning to be felt, leading to an increase in the use of Oth themes.
5. Process types
The categories of process types were outlined above. The distribution of the processes of finite verbs for the Mémoires is given in Table 8.
Table 8. Process types in the Mémoires.
| 1785 | 1835 | |||
| Mat | 323 | 53% | 181 | 40% |
| Ment | 71 | 12% | 63 | 14% |
| Rel | 183 | 30% | 170 | 38% |
| Verb | 22 | 4% | 29 | 6% |
| Exist | 7 | 1% | 10 | 2% |
| Total | 606 | 453 | ||
The percentage of material processes falls from 53% in 1785 to 40% in 1835 (seven percentage points). This is paralleled by a corresponding rise in the rate of relational processes, which rises from 30% in 1785 to 38% in 1835 (eight percentage points). Material processes encode actions and events of a physical nature, while relational processes encode static relationships, and hence state what is the case. So, there seems to be a movement from interest in dynamic physical happenings to the static nature of the world as it is. By 1835, interest in relational process has risen to be almost on a par with that for material process.
Table 9 gives the corresponding distribution for the Philosophical Transactions.
Table 9. Process types in the Philosophical Transactions.
| 1785 | 1835 | |||
| Mat | 268 | 47% | 261 | 28% |
| Ment | 100 | 18% | 181 | 20% |
| Rel | 176 | 31% | 385 | 42% |
| Verb | 20 | 4% | 78 | 9% |
| Exist | 7 | 1% | 12 | 1% |
| Total | 571 | 917 | ||
Here the same changes are in evidence, but the degree of the change is considerably greater than in the Mémoires. The percentage of material processes has decreased by 19 percentage points from 47% to 28%, while the rate of relational processes has increased by 11 percentage points from 31% to 42%. Hence, in the Philosophical Transactions there seems to be a relatively strong movement towards a static descriptive style, so that by 1835 relational process is clearly the dominant process type. It might also be noticed that the frequency of material process is less in the Philosophical Transactions than in the Mémoires at both of the dates: 6 percentage points in 1785 and 12 in 1835. At the same time the frequency of mental process is greater in the Philosophical Transactions than in the Mémoires: 6 percentage points at both dates. This seems to indicate that the writers of the Philosophical Transactions attach more importance to mental processes than their French counterparts do.
6. Final thoughts
If we consider the changes over time indicated by these results, a number of conclusions seem to follow. In the Mémoires, there is little change in the grammatical functions of themes, other than a small reduction in subject themes and rise in adjunct themes. In the Philosophical Transactions, there is also a reduction in subject themes and corresponding rise in adjunct themes, and the degree of change is greater than in the Mémoires. In the Mémoires there is hardly any change in the semantic categories of themes, other than a small increase in the Ment type. There is rather more change in the Philosophical Transactions, where we find a significant increase in the use of Obj themes, though rather less if the study group is taken as a whole. There is also a decrease in the use of Auth themes, but an increase in the use of Oth themes. In terms of process types, there is a decrease in the use of material process, and an increase in the use of relational process in the Mémoires. These same changes are present in the Philosophical Transactions, where the degree of change is considerably greater than in the Mémoires.
Without having a control group, and to the best of my knowledge there are no studies which could be used for comparative purposes, it is difficult to know whether these results would be typical of a 50-year period. Intuitively, however, I must admit the changes are smaller than I expected. Nevertheless, there seems to be a general pattern, whereby there are a number of limited changes in the Mémoires; these same changes appear in the Philosophical Transactions, where the degree of change is systematically greater than in the Mémoires. One can only speculate as to why this should be the case, but I would suggest that in times of great turbulence there is a tendency to be more conservative than would normally be the case. In times of danger, people hang on to whatever they can preserve. So, my suggestion is that in times of great difficulty, scientists attempt to continue doing things as they have done, changing as little as possible. If this hypothesis is correct, it would explain why the French scientists, having gone through the upheaval of the French Revolution, combined with foreign wars, changed their writing style very little. The English scientists had gone through a series of wars, notably those with Napoleon, but were not directly affected by the French Revolution, and consequently their writing displays a rather greater degree of change than that of their French counterparts.