Following the inspiration of Linnaeus, Guyton suggested in 1782 that chemical language should be based upon three principles: substances should have one fixed name; names ought to reflect composition when known (and if unknown, they should be non-committal); and names should generally be chosen from Greek and Latin roots and be euphonious with the French language. In 1787, Guyton, together with Lavoisier, Berthollet and Fourcroy, published the 300-page Méthode de nomenclature chimique, which appeared in English and German translations a year later. One-third of this book consisted of a dictionary, which enabled the reader to identify the new name of a substance from its older one. For example, ‘oil of vitriol’ became ‘sulphuric acid’ and its salts ‘sulphates’ instead of ‘vitriols’; ‘flowers of zinc’ became ‘zinc oxide’.
Perhaps the most significant assumption in the nomenclature was that substances that could not be decomposed were simple (i.e. elements), and that their names should form the basis of the entire nomenclature. Thus the elements oxygen and sulphur would combine to form either sulphurous or sulphuric acids depending on the quantity of oxygen combined. These acids when combined with metallic oxides would form the two groups of salts, sulphites and sulphates. In the case of what later became called hydrochloric acid, Lavoisier assumed that he was dealing with an oxide of an unknown element, murium. Because of some confusion over the differences between hypochlorous and hydrochloric acids, in Lavoisier’s nomenclature hydrochloric acid became muriatic acid and the future chlorine was ‘oxygenated muriatic acid’. The issue of whether the latter contained oxygen at all was to be the subject of fierce debate between Davy, Gay-Lussac and Berzelius during the three decades following Lavoisier’s death.
The French system also included suggestions by Hassenfratz and Adet for ways in which chemicals could be symbolized by geometrical patterns: elements were straight lines at various inclinations, metals were circles, alkalis were triangles. However, such symbols were inconvenient for printers and never became widely established; a more convenient system was to be devised by Berzelius a quarter of a century later.
During the eighteenth century some chemists had turned their minds to quantification and the possible role of mathematics in chemistry. On the whole, most chemists agreed with Macquer that chemistry was insufficiently advanced to be treated mathematically. Although he believed, correctly as it turned out, that the weight of bodies bore some relationship to chemical properties and reactions, the emphasis on affinity suggested that the project was hopeless. Nevertheless, Lavoisier, inspired by the writings of the philosopher, Condillac, believed fervently that algebra was the language to which scientific statements should aspire7:
We think only through the medium of words. Languages are true analytical methods. Algebra, which is adapted to its purpose in every species of expression, in the most simple, most exact, and best manner possible, is at the same time a language and an analytical method. The art of reasoning is nothing more than a language well arranged.
In a paper on the composition of water published in 1785, Lavoisier stressed that his work was based upon repeated measuring and weighing experiments ‘without which neither physics nor chemistry can any longer admit anything whatever’. Again, in another essay analysing the way metals dissolve in acids, Lavoisier used the Hassenfratz – Adet symbols:
In order to show at a glance the results of what happens in the solution of metals, I have constituted formulae of a kind that could at first be taken for algebraic formulae, but which do not have the same object and which do not derive from the same principles; we are still very far from being able to obtain mathematical precision in chemistry and therefore I beg you to consider the formulae that I am going to give you only as simple annotations, the object of which is to ease the workings of the mind.
The important point here was that Lavoisier used symbols to denote both constitution and quantity. Although he did not use an equals sign, he had effectively hit upon the idea of a chemical equation. As we shall see, once Berzelius’ symbols became firmly established in the 1830s, chemists began almost immediately to use equations to represent chemical reactions.
While producing the Méthode de nomenclature chimique with Lavoisier and the others, Guyton was converted to the new chemistry. Because the new language was also the vehicle of anti-phlogiston chemistry, it aroused much opposition. Nevertheless, through translation, it rapidly became and still remains the international language of chemistry.
TABLE 3.2 Lavoisier’s ‘elements’ or ‘simple substances’.
Lavoisier’s final piece of propaganda for the new chemistry was a textbook published in 1789 called Traité élémentaire de chimie (An Elementary Treatise on Chemistry). Together with Fourcroy’s larger text (published in 1801), this became a model for chemical instruction for several decades. In it Lavoisier defined the chemical element pragmatically and operationally as any substance that could not be analysed by chemical means. Such a definition was already a commonplace in mineralogical chemistry and metallurgy, where the analytical definition of simple substances had become the basis of mineralogical classification in the hands of J. H. Pott, A. F. Cronstedt and T. Bergman. It was for this reason that Lavoisier’s list of 33 basic substances bore some resemblance to the headings of the columns in traditional affinity tables. Lavoisier’s list included substances such as barytes, magnesia and silica, which later proved to be compound bodies.
After discussing the oxygen theory in part I of the Traité, he discussed their preparation and properties, their oxides and then their salts formed from acidic and basic oxides in part II. Caloric disengaged from oxygen explained the heat and light of combustion. It has been said that the elements formed the bricks while his new views on calcination and combustion formed the blueprint. The Traité itself formed a dualistic compositional edifice. Whenever an acidic earth and metal oxide (or earth) combined, they produced a salt, the oxygen they shared constituting a bond of union between them. As was appropriate for an elementary text, part III, a good third of the book, was devoted to chemical instrumentation and to the art of practical chemistry.
Lavoisier’s table of elements did not include the alkalis, soda and potash, even though these had not been decomposed. Why were they excluded from his pragmatic definition of simple substances? Two reasons have been suggested. In the first place, he was prepared to violate his criterion because of the chemical analogy between these two alkalis and ‘ammonia’, which Berthollet had decomposed into azote (nitrogen) and hydrogen in 1785. Lavoisier was so confident that soda and potash would be similarly decomposed into nitrogen and other unknown principles, that he withheld them from the table of simple substances. On the other hand, although confident that muriatic acid was also compound, because the evidence was not so strong as for the alkalis, he included it in the list of elements. While we may admire Lavoisier’s prescience – Davy was to decompose soda and potash in 1808 – this was a disturbing violation of his own pragmatism. What guarantee did the chemist have that any of Lavoisier’s simple substances were really simple? As we shall see, Lavoisier’s operational approach caused a century of uncertainty and helped to revive the fortunes of the ancient idea of primary matter.
A second explanation is more subtle. Lavoisier’s simple substances were arranged into four groups (see table 3.2). Three of the groups contained the six non-metals and seventeen metals then known, both of which were readily oxidizable and acidifiable, together with the group of five simple ‘earths’. The remaining group was light, caloric, oxygen, azote (nitrogen) and hydrogen. At first glance these elements appear to have nothing in common, but the heading Lavoisier gave them, ‘simple substances belonging to all the kingdoms of nature, which may be considered the elements of bodies’, provides the clue. Lavoisier probably saw these five elements as ‘principles’ that conveyed fundamental generic properties. Light was evidently a fundamental principle of vegetable chemistry; caloric was a principle of heat and expansibility; oxygen was the principle of acidity; hydrogen was the principle of water that played a fundamental role in all three kingdoms of Nature; and nitrogen was a principle of alkalinity.