The chief interest of the antiquarian in Rome’s ruins was topographical, identifying them if possible with structures known and described in Latin literature. Attempts to picture the layout of the ancient city generated numerous maps and disquisitions, which gradually morphed into guidebooks for tourists, many of which focussed on only the ancient remains to the exclusion of the modern city. The development of tourism is one of the capital outcomes of the fascination with the ruins of Rome. There does not seem to have been any other city or site in the world that was visited for the sake of its ruins. Topographical studies were, however, hampered by their reliance on more (or in one case, less) ancient texts in which buildings and their locations were mentioned, not always reliably. It became clear in due course that the only way forward lay in archaeological excavation.

Early states converge on a similar, number-based, “algorithmic” theoretical science. In Greek mathematics we see a new science, based not on the anonymous teacher but on the named author, seeking fame. Such authors look for new, surprising results, and therefore couch them in the language of proof. The resulting body of knowledge of many surprising proofs has no precedent in previous societies. The generation of Archimedes adds considerable subtlety and brings this to the cusp of modern science. The new Greek departure intertwines with other traditions, especially in the Near East, giving rise to a number-based but also proof-based astronomy (that of Ptolemy) and to the Arabic algebra. In the Renaissance, efforts to creatively engage with this Greek legacy gave rise to the scientific revolution. The science we know is objectively valid, but also historically contingent; one of the contingencies making it possible was the new departure of Greek mathematics.

Literary canons are neither the embodiment of transcendent values nor the artificial construct of gatekeepers. They are produced by deep social and poetic forces. The Greek emphasis on performativity meant that the new Athenian democracy became hyper-performative with its drama and speeches and outspoken figures such as Socrates. This riveted attention in the Greek world and, as Athenian political power disappeared, Athens’ soft power could not be replaced (since new sites of power were not as publicly performative). This created a Greek model of culture as a sphere autonomous from the state, predicated on the internal friction of the democratic city. This made possible a science based on debate. Replicated by later cultural traditions, it provided the model for Roman otium, for scientific debates from the Islamic world onward and for the rise of modern European civil society.

Petrarch initiated ruin-tourism, and that flowered in the period of the eighteenth-century Grand Tour. Arguably, the ruins of Rome were the first to generate the production of a considerable variety of souvenirs, portable objects manufactured expressly for visitors to take away. Now a souvenir is only desirable if the object it represents is deemed attractive: the ruin-aesthetic was so well established by the time of the Grand Tour that ruins moved from the background of paintings into the foreground; they became the subject. In the engravings of Piranesi the ruins of Rome reached their peak of aesthetic appeal. The aesthetic validation of ruins is to the fore, since the English decorated the interiors of their houses with scenes of ruination. They also brought home architectural models of ruins in cork or marble for display; their porcelain and fans were decorated with ruin motifs.

Interaction in Poetic Imagery was my first book. It was based on my doctoral researches at Cambridge in the mid-1960s. I set out to formulate a theory of a significant, but previously untheorised, aspect or potentiality of poetic imagery – indeed an aspect or potentiality not previously demarcated, either under the name I gave it, ‘interaction’, or any other name. With ‘imagery’ understood as ‘metaphor, simile and the various forms of comparatio’, I identified interaction as ‘any local cross-terminological relation between the tenor and vehicle of an image’, explicitly adapting ‘tenor’ and ‘vehicle’ from I. A. Richards’ The Philosophy of Rhetoric. I then categorised the modes of interaction into four groups (with subsequent subdivisions). The cross-terminological relation might be effected aurally or extra-grammatically or by intrusion or (most commonly) by neutral terminology. All this was established on the basis of a corpus of Greek poetry – early lyric and dramatic poetry, from Archilochus to Aeschylus – with additional examples from English poetry, from Shakespeare to the twentieth century.

I have reserved for discussion here certain passages containing more than a single interaction. Not all combinations are significant, though. The significant combinations in the Greek corpus are of several kinds, of which the best attested and conceptually simplest comprises those instances with a broadly or cumulatively preparatory effect in favour of the vehicle.