TO THE CASSITERIDES AND BEYOND: GEORGE BONSOR’S ARCHAEOLOGICAL EXPLORATION OF THE SCILLY ISLES (1899–1902). AN HISTORIOGRAPHIC OBLIVION

Abstract

Between 1899 and 1902, Anglo-French archaeologist George Bonsor carried out an exploration of the Scilly Isles (United Kingdom). At that time the archipelago was believed to be the Cassiterides or Tin Islands mentioned by authors such as Strabo, Pliny the Elder and Ptolemy – an idea first posited by William Camden in his Britannia (1586). Adopting Camden’s theory and guided by ancient literature on the Cassiterides – which refers to the Phoenicians as the first controllers of this trade route – Bonsor sought traces of the Phoenicians and their tin trade in the Scilly Isles, becoming the first person to conduct such research from an archaeological perspective. Not having found any evidence, his exploration remained unpublished and went mostly unnoticed in debates about the Tin Islands over following decades. This paper presents a brief historiographical account on the Cassiterides before and after the explorations, as well as a critical analysis of Bonsor’s field notes regarding his use of ancient sources and his archaeological method. The analysis carried out suggests that Bonsor’s archaeological exploration has been overlooked thus far and that a new assessment of his work is required.

INTRODUCTION

The Cassiterides, or ‘Tin Islands’, are an archipelago mentioned by ancient sources from Herodotus (fifth century bc) to Solinus (fourth century ad) as being rich in tin, just as their toponym, from the Greek, κασσίτερος (‘tin’) might suggest. Since the sixteenth century the geographical location referred to by this place-name has been debated from the perspectives of ancient literature and historiography, and from the end of the nineteenth century by archaeology. This paper addresses all three perspectives, but focuses chiefly on more modern discussions, including the first archaeological attempt to locate the Cassiterides by George Bonsor (1899–1902). The main argument of this paper is that Bonsor’s archaeological exploration of the Scilly Isles has been inaccurately assessed in the context of the research on the Tin Islands.

Ancient literature on the archipelago includes at least nine testimonies from seven different writers, including Herodotus.¹ Of these authors, several have been more concerned in identifying the Cassiterides: Strabo, Pliny and, indirectly, through his influence in modern cartography, Ptolemy.² The reference by Strabo of Amaseia (64 or 63 bc–c ad 24) in his ‘Geography’ is undoubtedly the most exhaustive source about the islands. The author refers to the Cassiterides as islands located near the north-western coast of Iberia, postulating that:

The Cassiterides are ten in number, and lie near each other in the ocean towards the north from the haven of the Artabri. One of them is desert, but the others are inhabited by men in black cloaks, clad in tunics reaching to the feet, girt about the breast, and walking with staves, thus resembling the Furies we see in tragic representations. They subsist by their cattle, leading for the most part a wandering life. Of the metals they have tin and lead; which with skins they barter with the merchants for earthenware, salt, and brazen vessels. Formerly the Phoenicians alone carried on this traffic from Gades, concealing the passage from every one; and when the Romans followed a certain ship-master, that they also might find the market, the shipmaster of jealousy purposely ran his vessel upon a shoal, leading on those who followed him into the same destructive disaster; he himself escaped by means of a fragment of the ship, and received from the state the value of the cargo he had lost. The Romans nevertheless by frequent efforts discovered the passage, and as soon as Publius Crassus, passing over to them, perceived that the metals were dug out at a little depth, and that the men were peaceably disposed, he declared it to those who already wished to traffic in this sea for profit, although the passage was longer than that to Britain.³

Strabo’s description constitutes the most detailed testimony about the Tin Islands,⁴ and the passage has accordingly played a central role in modern and contemporary discussions about this matter from the perspective of ancient literature. The geographer states that the Cassiterides are a group of ten islands located northward from the ‘haven of the Artabri’, ie Galicia in Spain. Furthermore, he clearly states that the tin trade from them was originally under Phoenician control. The quality of being islands, the control of the tin trade there by Phoenicians and the quest for prestige were the main reasons that historians of the sixteenth to nineteenth centuries laid territorial claim to the islands on behalf of their nations.⁵ A pioneering figure to do so was William Camden (1551–1623), who carried out the first systematic work on the Cassiterides. On the recommendation of Abraham Ortelius,⁶ Camden thoroughly re-examined the ancient sources and published his interpretation in the monumental Britannia (1586), concluding that the Tin Islands of antiquity were in fact the Scilly Isles. It was this work that later led Bonsor to explore the archipelago to try and confirm Camden’s hypothesis.

Conceived originally in Latin as a project for the Crown of England, Britannia was translated into English and published in different editions up to the early eighteenth century.⁷ It consisted of a geographical-historical treaty that followed the chorographical tradition dating back to the Italian Quattrocento.

In order to contextualise Camden’s work, it is crucial to understand the practice of ‘chorography’, deriving from the Greek: χῶρος (‘place’, ‘space’) + γραφια (‘writing’). Chorography is the act of describing or mapping a place, often by combining historical narratives, geographical descriptions and the accounts of classic authors such as Pliny, Ptolemy or Strabo, whose works deal with ancient geographical knowledge.⁸ Therefore, its objective is to link ancient descriptions to modern places.

Based on the solid hermeneutics of ancient authors, Camden’s chorographical work was the most influential in locating the Cassiterides islands from the sixteenth century onwards. In the 1722 edition of Britannia, the zone between Cornwall and the Scilly Isles is described by Camden:

… and from this place the tide presses both to the north and east, with great noise and violence, being streighten’d between Cornwall and those islands which Antoninus calls Sigdeles, Sulpitius Sillinae, Solinus Silures, the English Silly, the Dutch seamen Sorlings, and the ancient Greeks Hesperides and Cassiterides.⁹

In addition to demonstrating a wide knowledge of the ancient sources referred to, the author analyses Strabo’s prominent testimony¹⁰ in detail to reach his conclusion:

Now, considering that these Isles of Scilly are opposite to the Artabri, i.e. Gallitia, in Spain; that they stand directly north of them; that they lie in the same Climate with Britain; that they face Celtiberia; that the Sea is much broader between them and Spain than between them and Britain; that they lie just upon the Iberian Sea, and close to one another, northward; that there are only ten of any note […]. From so many concurring testimonies, I should rather conclude these to be the Cassiterides […].¹¹

Camden then proceeds to list the theories that he knows about, but does not accept: ‘the Azores which lie too far westward, or Cisarga (with Olivarius) which in a manner joins to Spain; or even Britain it self, with Ortelius; since there were many of the Cassiterides.’¹² Unsurprisingly, Camden’s interpretation remained unquestioned until the end of the eighteenth century, when two Spanish authors – José Cornide and Miguel Ignacio Pérez – proposed alternative views in nationalistic reaction to his statements.¹³ These authors meant to ‘restore’ the Tin Islands to Galicia, a hypothesis that was little-known outside their region.¹⁴ Two factors may explain the varying degrees by which the two perspectives were extended. In the case of Camden, there was the influence of his fame, and the fact that Britannia was part of a royal initiative. Cornide and Pérez, on the other hand, were only well known in Spain, and theirs were personal projects. In fact, along with Leland and Lambarde, Camden was considered a key figure of chorographic scholarship in the sixteenth century.¹⁵ His work not only emphasised the importance of local history, but also had a clear underlying nationalistic purpose promoting the idea of an independent national past built by the Tudors¹⁶ – a notion that gave his work particular relevance within Britain.

It was not until the very end of the nineteenth century that someone decided to try and put over three centuries of writings about the Tin Islands to the test. George Bonsor Saint-Martin (1855–1930)¹⁷ was an Anglo-French artist (his father was English while his mother was French) who moved to Spain and, having observed ancient Roman remains at Carmona (Seville), progressively gained an interest in archaeology. By the last decades of the nineteenth century, he had already earned recognition within part of archaeological academia of the time. Having found traces of Phoenician population in Los Alcores,¹⁸ he intended to follow their trail towards the famous Tin Islands.

Bonsor’s findings were detailed in two manuscripts (corresponding to 1899–1900 and 1901–2 respectively), currently in the Archivo General de Andalucía (AGA) and in the process of being published. Although these documents are known in the context of the history of the Scilly Isles,¹⁹ they have not been previously analysed with regard to discussions about the Cassiterides. In order to carry out such analysis, the methodology used in this work consists of three main sections: first, an historiographic account of the Cassiterides between the sixteenth and the twentieth centuries; second, a study of the primary sources that have been transcribed using palaeographical methods;²⁰ and third, a critical analysis of Bonsor’s archaeological practice, including his use of ancient sources, in the context of the late nineteenth-century discipline.²¹ Some figures have been included to highlight Bonsor’s expertise at documenting his archaeological findings. In these cases, relevant notes have been translated into English by the author.

WITH PHOENICIANS IN MIND: ON THE RELEVANCE OF THE EXPLORATION OF THE SCILLY ISLES IN NINETEENTH-CENTURY EUROPE

Among English people of the nineteenth century there was a widely held belief that the ancient Phoenicians had reached the English coasts (Cornwall and the Scilly Isles) to trade for tin in the Cassiterides.²² This belief was largely (but not exclusively) a reflection of the influence that Camden’s ideas had for over three centuries, enhanced by their publication in a geographical and historical work promoted by Elizabeth I (r 1558–88). Taken together, Camden’s long-term reach and the sense of historical singularity underlying the creation of the English nation would popularise the idea that the English people somehow resembled the Phoenicians in their ability to carry out epic enterprises. It helped to justify the ‘civilising’ mission they were developing at the time, along with other European countries, ie building their colonial empire. All the above consequently underscored the positive perception of the Phoenicians, who were supposed to have played the same role of spreading civilisation during antiquity.²³

The alleged Phoenician tin trade in the Scilly Isles is well represented in history and art from the seventeenth century onwards. Besides Samuel Bochart,²⁴ who supported the idea, Aylett Sammes,²⁵ following the latter’s work, even proposed that the toponym Britannia had a Semitic origin and made Himilco/Herakles the discoverer of Britain.²⁶ The same is true for Joseph Haydn’s Dictionary of Dates, where the Scilly Isles entry is headed by the equation between the British and the tin-bearing archipelagos.²⁷ The painting by Frederic Leighton, Phoenicians bartering with ancient Britons (1894–5), gives a very accurate depiction of how this trade was envisioned by contemporary England, and so do historians such as John Whittaker and George Rawlinson. The latter were sources of Bonsor’s that exemplify how firm Camden’s theory remained. In the book by Richard Polwhele, History of Cornwall,²⁸ edited by Samuel Drew, Whittaker states the following:

They were Phenicians [sic] transplanted to Cartaghe in Africa, and again transplanted to Cadiz in Spain. From their settlement at the latter, inheriting all the nautical genius from their Tyrian ancestors, and improving it in adventures upon the once dreaded Atlantic before them; witth a spirit of enterprise, which reflects high honour upon them, they found their way to the Sylley Isles at the nearest end of our own Britain.²⁹

On the one hand, this passage reflects the assumed role of the Scilly Isles in the tin trade and, on the other hand, the way the British self-identification with the Phoenicians configured a positive perception of the latter (see emphasis).

While working in Los Alcores, Bonsor’s aim was to study the Phoenician population settled there from about the ninth century bc. The varying interactions between Phoenicians and local peoples of the Iberian Peninsula, depicted by ancient authors and represented in the archaeological record, is currently believed to be the origin of the Tartessos civilisation.³⁰ Bonsor’s extension of this research led him, following Camden, to explore the question of the Cassiterides.

It should be noted that Bonsor’s manuscript of 1899–1902 does not explicitly mention Camden (he is first cited in a later work ‘Tartessos’³¹), although it does refer to the supposed ancient names for the Scilly Isles noted in the 1722 edition of Britannia.³² This reference reinforces the continuing influence of Camden’s hypothesis, certainly in British circles, but also beyond. Already by the end of the eighteenth century, authors such as Cornide and Pérez were willing to deconstruct the theory, and yet it continued to hold sway among Bonsor’s contemporaries, Federico Maciñeira and Luis Siret.³³ In Spain, Camden’s ideas were supported by authors such as Juan Francisco Masdeu and Luis José Velázquez, 2nd Marquis of Valdeflores.³⁴ They also found advocates in nineteenth-century Germany, where they were accepted by Alfred Gutschmid, Theodor Mommsen and Emil Hübner (see below).

Set against this scholarly background there was some impetus behind the archaeological exploration of the Scilly Isles, especially from Britain, but also other European countries such as Spain, Italy and Germany. By the end of the nineteenth century there were high expectations for identifying scientific proof of Phoenician activity on the Scilly Isles. When this evidence failed to materialise, the enthusiasm for archaeological exploration unsurprisingly waned.

REASSESSING THE ARCHAEOLOGICAL EXPLORATION OF THE SCILLY ISLES

The reception of the archaeological exploration of the Scilly Isles

After the exploration of the Scilly Isles (1899–1902) ended, a myriad of theories proliferated over where the Tin Islands might be located. In Great Britain, Thomas Rice Holmes compiled a list of the theories circulating at the beginning of the twentieth century,³⁵ seemingly unaware of George Bonsor’s exploration. Some of the most relevant proposals were the following: authors such as Samuel Bochart, Salomon Reinach and Karl Müllenhoff, as well as Rice Holmes himself, identified the tin-bearing islands with the British Isles.³⁶ Both Samuel Bochart³⁷ and later Henri D’Arbois de Jubainville championed the idea that Pliny’s Midacritus, the first person ‘to ever bring plomb from the Cassiteris island’,³⁸ should be identified with the Phoenician deity Melqart and considered an incarnation of their race.³⁹

Another group of academics championed the idea that the Cassiterides were to be found in Galicia, Spain. This included, among others, M Hans Hildebrand and Francis Haverfield,⁴⁰ which is surprising considering the fact that Cornide’s and Pérez’s writings, not particularly well known outside their country, had mainly been driven by their will to cancel Camden’s hypothesis for the sake of their homeland.

Lastly, the idea that the Scilly Islands were the Tin Islands of antiquity was still maintained, almost contemporaneously to George Bonsor’s exploration, by prominent figures such as Theodor Mommsen and Emil Hübner,⁴¹ although they barely referred to the matter, perhaps reflecting how entrenched the theory was in general perception.

In the Iberian Peninsula, Luis Monteagudo gave another account of the different trends.⁴² Along with Jorge Maier,⁴³ Monteagudo was one of the few authors aware of Bonsor’s campaigns in the archipelago. Monteagudo argued that the toponym ‘Cassiterides’ had been applied in early sources to every relevant tin-bearing region, and drew attention to the direction and order in which the names, as recorded, may have referred to each other. So, according to Bonsor as well as authors such as Obermaier and Serpa Pinto, the displacement took place from north to south in three stages between Galicia and the nearby isles (sixth century bc, until the battle of Alalia), Armorica and, from Roman times, Great Britain; whereas other academics, such as Adolph Schulten or Luis Siret, proposed the opposite direction, from south to north, with Galicia being last in the sequence of referends.⁴⁴

The fact that only Monteagudo referred to Bonsor’s explorations highlights the extent to which his campaigns in the Scilly Islands went unnoticed in discussions about the location of the Tin Islands. This is surprising given its potential impacts on the historiographical debate, both for the study of the early tin trade and the archaeology of the Scilly Islands more generally. Regarding the former, even though no traces of Phoenician presence were found in the archipelago, the campaigns were still interesting to contemporary archaeologists such as Federico Maciñeira, as shown by Bonsor’s correspondence:

Me extraña que en sus importantes exploraciones de las Islas Sorlingas no encontrase rastro de las gentes púnicas, […] porque, aún dando por hecho que no sean las Cassitérides, al fin por su proximidad a Cornuallia y por el estaño que algunos dicen que abunda es lógico suponer que no dejasen de beneficiarlas aquellas gentes.⁴⁵

Regarding the archaeology of the Scilly Isles, Bonsor was repeatedly requested to publish his findings, first by Reginald A Smith, Keeper of the Department of British and Medieval Antiquities at the British Museum, and later by Sir Thomas Kendrick, director of the British Museum at the time, on behalf of Hugh O’Neill Hencken.⁴⁶ Smith offered to publish Bonsor’s research in The Antiquaries Journal and Bonsor accepted, but never got around to writing it up.⁴⁷ Instead, Bonsor sent Kendrick copies of a part of his work in 1927, including most of the figures published by Hencken a few years later.⁴⁸ All these approaches to Bonsor provide clear evidence of the interest fellow contemporary researchers had in his work, even though he was unsuccessful in confirming the identity of the Cassiterides.

In sum, it is clear from Luis Monteagudo’s essay that Bonsor’s work remained unknown to most scholars at least until the late 1950s. The only paper published on the issue by Bonsor appeared in 1928, and this only noted the absence in the archaeological record of a Phoenician presence in the Scilly Isles. The following section aims to reassess the work carried out by Bonsor in the Scilly Isles.

Towards a reassessment of the archaeological exploration of the Scilly Isles

One aspect that distinguishes Bonsor’s archaeological exploration from others of his day was the variety of methodologies that he utilised, drawing as it did on philology, fieldwork and – as the campaigns proceeded – a linguistic-cultural method.

Bonsor’s textual research drew upon an up-to-date knowledge of written historical sources. This is evident from the texts he quotes during the first two campaigns (1899–1900), which fall into different categories. Bonsor’s critical analysis of these texts is revealing.

The first group of texts are those that provide general information about the Tin Islands and the tin trade. Among these is Pliny the Elder’s (ad 23–79), Naturalis historia, in which Book 34 Chapter 16 states that ‘The tin was called Cassiteron by the Greeks, and fabulously narrated to be sought in islands of the Atlantic sea, and to be bought to the seekers in wicker boats sawed round with leather’⁴⁹ – a story that the author would appear to have found somewhat fantastical.

Another account Strabo attributes to Posidonius, saying that the tin is carried from the British Isles to Marseilles.⁵⁰ However, Strabo’s most relevant passage on the Cassiterides appears within a description of Iberia and the adjacent islands, cited above,⁵¹ in which the geographer would seem to denote Iberia, rather than the hypothesis Bonsor favoured. Despite this, Bonsor was keen to use Strabo and Pliny’s accounts to prove his theory. Strabo’s is the most detailed account, and Pliny has added more specific spatial details.

A second group of sources name particular individuals taking part in the tin trade, either as being the first to have brought tin from the islands (ie Pliny’s Midacritus) or for being responsible for the Roman discovery of the route (ie by Publius Crassus). The identity of both individuals is subject to debate. Midacritus is a hapax, while Publius Crassus could refer, with Roman naming conventions, either to the triumvir Marcus Licinius Crassus (115–53 bc), or to his father Publius Licinius Crassus (consul 97 bc). Regarding the latter, Pliny states that the Greeks of Marseilles, along with the Phoenicians, exported tin from the Cassiterides. Bonsor suggests that this must have taken place prior to Caesar’s arrival in Britain,⁵² which would seem to identify the triumvir’s father as the discoverer of the route. Depending on where the source of tin was, this might be taken to reflect Roman interests either in the metalliferous zone of northern Iberia⁵³ or, as Bonsor surmised, the British Isles as the core territory for the tin trade.

A third group of sources comprise those by authors whose testimony, for different reasons, are not helpful for identifying or locating the Tin Islands. One of these is Herodotus,⁵⁴ whose idea of a western archipelago from which tin was brought is brief and lacking in detail.⁵⁵ Similarly, Rufus Festus Avienus, writing in the fourth century ad, refers to the ‘Oestrymnides’ islands, apparently the Cassiterides, but the composition of his ‘Ora Maritima’ does not sit well, in terms of homogeneity and coherence, with other Greek periploi of the time.⁵⁶

In conclusion, it is fair to claim that Bonsor’s textual research took account of the most relevant ancient sources from Herodotus to Pliny, even if these were used somewhat uncritically. In this, he was little different from most scholars of the late nineteenth and early twentieth centuries, who saw ancient Graeco-Roman authors as the departure point for any serious historical or even archaeological research. Nevertheless, his reading of the evidence was heavily influenced by Camden’s work.

Despite how firmly established the assumption was that the Scilly Isles were the Tin Islands, it was Bonsor’s belief that only archaeology could confirm or refute this identification. To address the question of the Cassiterides from a material point of view, Bonsor undertook a large programme of fieldwork, including survey, recording and targeted excavation. His investigations started in 1899 with some surveys and the first plans. In 1900 he carried out excavations at Halangy tomb (now known as Bant’s Carn burial chamber) and Halangy Porth village, both at the northern tip of the island of St Mary’s.

Through his fieldwork, Bonsor acquired significant knowledge of the archaeology of the archipelago. The two excavations, in particular, show his ability to create high-quality documents in which his findings were reflected and clearly placed in context. His method favoured drawing as a fundamental tool to document his findings, combined with precise measurements and measured annotations. Two section drawings of the stratigraphic sequences are examples of his work (figs 1 and 2). The first features short descriptions of each layer and the second uses letters to identify the different elements found.

Fig 1. First stratigraphy of Hallangy Porth village. Bonsor 1899–1900, 130. Image: Reproduced with permission from AGA. See also Maier 1999a, 153, fig 45.

Fig 2. Second stratigraphy of Hallangy Porth village. Bonsor 1899–1900, 141. Image: Reproduced with permission from AGA. See also Ashbee 1974, 161, fig 30; Maier 1999a, 154, fig 46.

Some of his drawings show an artistic concern, reflecting a common approach of the time, which R Bianchi Bandinelli has called ‘philological archaeology’.⁵⁷ The term encompasses the main branch of classical archaeology in the nineteenth century, defined by the prominence of classical texts as the source for the discovery of ancient monuments and/or places, with special attention given to epigraphs and coins.⁵⁸ Another key feature of this perspective is the imposed hierarchy of objects based on their artistic grandeur (ie their degree of beauty and perfection) according to Graeco-Roman standards. This idea derives from J J Winckelmann’s concepts of ancient art, articulated in the second half of the eighteenth century.⁵⁹ Given these influences in Bonsor’s time, it is unsurprising that the artistic qualities of drawings were often regarded more important than their analytical ones. Even so, the quality of Bonsor’s drawings is noteworthy:

This plan and section are remarkable because such drawings have only recently become the rule, rather than the exception, and they show, at the beginning of this century, when often the standards of the excavation left much to be desired, an acuteness of perception, and competence considerably in advance of the time.⁶⁰

In addition to the section drawings described above, Bonsor also recorded several kistvaens, or entrance graves (figs 3 and 4).⁶¹ These structures are tombs or burial chambers with an entrance formed from flat stone slabs in a box-like shape, sometimes covered by a tumuli or barrow.⁶² Bonsor recorded these structures both in plan and section, sometimes limiting the depiction to their funeral chamber while in other cases including the tumuli as well.

Two observations are worth emphasising regarding Bonsor’s field documents. The first, echoing Ashbee’s remarks quoted above, is the quality of Bonsor’s drawing when compared to that of contemporary archaeologists – in fact, Ashbee has pointed out that Bonsor’s plans and sections were used in 1970 to aid the restoration of Bant’s Carn burial chamber.⁶³ The second is the quality of the field research itself.

There is an analytical accuracy to the plans and sections that Bonsor made; measurements are indicated either in metres (m) or in feet (ft) and inches (in.). He also consistently gives orientations in the form ‘funeral chamber northeast–southwest’. The clear and consistent graphic code the author uses allows for a simple interpretation of these drawings. This is the case for covering and background stones in fig 4, represented in dark to signify those cut by the longitudinal section line shown in fig 3. Moreover, any elements missing or reconstructed (usually covering stones or parts of the circles of stones) are represented using discontinuous lines, which occurs in the case of the ancient huts of the Scilly Isles (fig 5). Notice the difference between the metric measurements indicated in figs 3 and 4 using a 5m scale and those annotated in fig 5 in imperial measurements – for example, the inner circle of one of the ‘British huts’ at Puckie’s Carn measures 14ft (4.27m).

Fig 3. Plan of the funeral chamber of the kistvaen known as Bant’s Carn or Hallangy tomb. Bonsor 1899–1900, np. Image: Reproduced with permission from AGA. See also Hencken 1932, 22–4, fig 10; O’Neil 1949, 15, fig 3; Ashbee 1974, 81, fig 10; Maier 1999a, 155, fig 47.

Fig 4. Longitudinal section of the kistvaen known as Bant’s Carn or Hallangy tomb. Bonsor 1899–1900, 163 (3). Image: Reproduced with permission from AGA. See also Hencken 1932, 21, fig 8; Ashbee 1974, 82, fig 11, and 1976a, 13, fig 2; Maier 1999a, 155, fig 47.

Fig 5. Plan depicting the remains of ‘British Huts’ at Bryher, Scilly Isles, with its shape reconstructed at the bottom right corner. Bonsor 1899–1900, 35. Image: Reproduced with permission from AGA.

These aspects are key to understanding Bonsor’s fieldwork.⁶⁴ According to his notes, Bonsor carried out his excavations at Halangy tomb and Halangy Porth village between September and early October 1900.⁶⁵ The excavation yielded chiefly handmade black pottery (decorated and plain) with quartz inclusions,⁶⁶ both from the tomb itself (fig 10) and from the midden shown in fig 2 (B), which Bonsor studied in more detail.⁶⁷ Indeed, it was the identification of the same pottery from both sites that led him to conclude that they were contemporaneous.⁶⁸ Besides that pottery, Bonsor found a ground stone and fragments of others, part of a stone axe, fragments of black flint, burnt stones and animal bones of different sizes.⁶⁹ Photographs of the tomb, currently part of the Lord Onslow Album (nos 29, 30 and 31), were taken by Alexander Gibson at some point after the excavation, as described by Ashbee.⁷⁰ As for the huts, fig 5 shows what the archaeologist referred to as ‘indications’ for them, but noted that one still had two large entrance stones, as can be seen top right (fig 5).⁷¹

An example of Bonsor’s desire to depict in detail both plans and sections of the most prominent structural remains he found are the figures he recorded of Innisidgen kistvaen, or Giant’s Grave (figs 6–8). The structure was possibly excavated at some point before the plan and sections were drawn,⁷² but Bonsor does not give any account of this in his notes. However, he did remark on the similarity between this megalithic monument and Halangy tomb,⁷³ both in terms of stone selection and architecture.⁷⁴

Fig 6. Plan of Giant’s Grave, also known as Innisidgen kistvaen. Bonsor 1901–2, np. Image: Reproduced with permission from AGA. See also Hencken 1933, 18, fig 5; O’Neil 1949, 14, fig 2; Daniel 1950, 62, fig 10; Ashbee 1974, 81, fig 10; Maier 1999a, 142, fig 42.

Fig 7. Longitudinal section of Giant’s Grave or Innisidgen kistvaen. Bonsor 1901–2, np. Image: Reproduced with permission from AGA. See also Hencken 1933, 18, fig 5; Daniel 1950, 62, fig 10; Ashbee 1974, 81, fig 10; Maier 1999a, 142, fig 42.

Fig 8. Cross-section of Giant’s Grave or Innisidgen kistvaen, taken from the middle diameter of the tumulus. Bonsor 1901–2, np. Image: Reproduced with permission from AGA. See also Hencken 1933, 18, fig 5; O’Neil 1949, 14, fig 2; Daniel 1950, 62, fig 10; Ashbee 1974, 81, fig 10; Maier 1999a, 142, fig 42.

Paul Ashbee’s detailed account of Bonsor’s fieldwork in the Isles of Scilly focused on a different megalithic structure: the so-called Grand Barrow or Great Tomb on Porth Hellick Down,⁷⁵ which the author implies was excavated in 1899. However, Bonsor’s notes clearly state that the excavation started in September 1902,⁷⁶ and there is no evidence whatsoever that he had undertaken any previous excavation there, although he had examined it several times before.⁷⁷ In all other respects Ashbee’s account of the excavation is clear enough to require no further comment.⁷⁸

Lastly, Ashbee refers to Obadiah’s Barrow, an important structure on Gugh, St Agnes, excavated by Bonsor in 1901 (fig 9).⁷⁹ This excavation is arguably one of the clearest examples of his field methodology, demonstrating an acuteness of perception and the care with which Bonsor carried out this work.⁸⁰ According to his notes, the excavation took place in September 1901.⁸¹ Bonsor provides an account of the deposits in the chamber and a plan and section of it (fig 10). The latter includes numbers and letters to refer to the covering stones and the distribution of finds along the chamber. Among these were pottery sherds of two kinds (black and reddish), several urns and sherds, some of them with burnt human bones, bone points and a scrap of bronze (fig 11) and ashes scattered, which made Bonsor think that the urns had been emptied at some point. Notably, he was also interested in ascertaining whether or not the bones he found at different times during this excavation belonged to the same individual.⁸²

Fig 9. Section of the funerary chamber of Obadiah’s Barrow. Bonsor 1901–2, 194. Image: Reproduced with permission from AGA. See also Ashbee 1980, 61, fig 4.

Fig 10. Plan and entrance section of Obadiah’s Barrow. Bonsor 1901–2, 205. Image: Reproduced with permission from AGA. See also Hencken 1932, 27, fig 12; Ashbee 1974, 93, fig 14.

Fig 11. Tips probably made of cervid horn and a scrap of bronze found at Obadiah’s Barrow. Bonsor 1901–2, 219. Image: Reproduced with permission from AGA. See also Hencken 1932, 27, fig 12; Hencken 1933, 24, fig 9b; Ashbee 1974, 240, fig 49.

Bonsor’s excavations in the Scilly Isles were by no means limited to those described above, but these examples are quite telling in showing the evolution of his methodology, both in terms of the analysis of the evidence he found and his documentation process, generating an archive that has proven useful for researchers many decades later.

Besides the drawings he made of stone monuments and structures, Bonsor also recorded bone and stone tools (fig 11) and, of course, pottery (fig 12). In this way he demonstrated an appreciation of the value that more mundane objects might have in understanding archaeological sites.⁸³ This aspect, exemplified by the diversity of the remains he documented, strengthens his relevance within the development of modern archaeological thought.

Fig 12. Sample of decorated pottery recovered at Hallangy tomb. Bonsor 1899–1900, 152. Image: Reproduced with permission from AGA. See also Hencken 1932, 23, fig 10; Ashbee 1974, 253, fig 56.

Bonsor had a keen interest in the recovered pottery. He was careful to record their decoration, which was reproduced in his drawings along with a description of the decorative techniques,⁸⁴ and also endeavoured to analyse the variety of clays used in their manufacture.⁸⁵ To this end, Bonsor exchanged letters with John J Harris, a Scillonian specialist in ceramics, which are available at the ‘Epistolario de Jorge Bonsor’,⁸⁶ a compilation of the letters sent and received by him, alongside the letters contained in the manuscripts analysed here.

Another noteworthy aspect of Bonsor’s work was his attempt to describe the places he studied in terms of how they may have been perceived by past societies. A very early instance of this can be found in 1899, when he notes that the place called Porth Hellick should have been the ancient, pre-Roman port, based on its landscape setting.⁸⁷ He subsequently developed this idea further by trying to reconstruct the ancient coastline of the island of St Mary’s.⁸⁸

Bonsor was also an early champion of archaeological photography.⁸⁹ He was keenly aware of its potential for documenting sites from a naturalistic perspective, which, allied with his detailed analytical drawings, could comprise a more complete record of the archaeological evidence. In this context, the photographs taken by Alexander Gibson and his family are key to a deeper understanding of Bonsor’s work, recording the scale and natural context of sites in detail.⁹⁰ Proof of the association between Gibson and Bonsor can not only be found in the photographs, but is also evident in Bonsor’s notes.⁹¹

Every aspect of Bonsor’s approach is more remarkable when taking into consideration that, particularly until 1904, he financed most of his own archaeological activity, unlike many other renowned archaeologists of the time.⁹² Even so, his research design had to be modified when no traces of a Phoenician presence in the Scilly Isles were identified. He returned to the linguistic debates about the name Cassiterides and the ethnicity of the population that carried out its tin trade. As shown in his diaries, it was after reading Salomon Reinach’s work, ‘L’étain celtique’,⁹³ that he began to reassess his previous conclusions. Salomon Reinach (1858–1932) was an eminent French historian, philologist and archaeologist. He proposed that the tin trade of the Cassiterides islands was first carried out by the Celts rather than Phoenicians, arguing that the name ‘Cassiterides’ had a Celtic origin, and therefore probably denoted the British Isles, and that Pliny’s Midacritus should instead be read as ‘Midas Phryx’.⁹⁴

In light of the new evidence, Bonsor changed his initial hypothesis, arguing instead that the Phrygians,⁹⁵ not the Phoenicians, had discovered the route of tin and brought it to the Mediterranean, with the latter only benefitting from it afterwards. Even though the archaeological campaigns continued, the question of the Cassiterides seemed partially answered. By matching the archaeological evidence to new philological thinking, he could reconcile the theory proposed by Camden.

As far as Reinach is concerned, his ideas about the tin trade would effectively displace the Phoenicians from the important role, following Strabo, that they supposedly played. Reinach’s thesis nevertheless requires some context. Bernal has pointed out that Reinach was an assimilated Jew who, despite his pro-Jewish attitude during a period of rising anti-Semitism, could have intended to differentiate himself – as a European Jew – from the Semitic people of antiquity, that is, the Phoenicians and Punics.⁹⁶ No similar intention can be gleaned from Bonsor’s manuscripts; he appears simply to have accepted Reinach’s work as research data. As much seems to be suggested by the evolution of his thinking about the Tin Islands, as described in his ‘From Tarshish to the Isles of Tin’.⁹⁷

On this point, Bonsor’s theoretical frame of reference deserves some reflection, since it naturally underpins the archaeological explorations analysed here. Besides being influenced by ancient texts, Bonsor was a convinced diffusionist, which had major implications for his way of interpreting the archaeological evidence.⁹⁸ According to Bonsor, Camden’s identification of the Tin Islands as the Scilly Isles did not in fact require revision, only the agents involved in the tin trade. Since the Cassiterides were known to be the centre of the tin trade, they would continue to be so regardless of who carried out this trade. This was also a comfortable position for Salomon Reinach, one of whose main concerns was to prove that ‘civilisation’ had a European origin, spreading from west to east, rather than the other way around.⁹⁹ His theory, and Bonsor’s support of it, allowed Reinach to downplay the influence of Semitic people in antiquity, including their role in the ancient tin trade.¹⁰⁰

CONCLUDING REMARKS

All in all, the archaeological explorations described in this paper did not result in a significant change in the European archaeological thinking. Nevertheless, they present an original application of it within studies dominated by textual approaches. Moreover, they contributed to wider discussions about the Phoenician presence in the west and, more fundamentally, to questions about the origin of ‘civilisation’. Such debates owe much to the archaeological exploration of the Scilly Isles.

Whatever the merits of Bonsor’s argument about the Cassiterides, there is little doubt that his work was pioneering in displacing classical authors by archaeology as the main source for the study of the ancient tin trade. Until 1899, archaeology had never been considered the right tool to approach this question. Thereafter, the matter would also need to be addressed from an archaeological perspective.

One last aspect to comment on is the extent to which Bonsor’s research remained valid after it took place. Certainly, he generated an interest in the archaeological potential of the Scilly Isles. He also produced a legacy for the archaeological work that came after his. Bonsor’s excavations at the sites of Halangy Down and Porth were resumed by Ashbee in 1964 and continued for over ten years, yielding discoveries of a further stone building, and pottery, including a sherd resembling Beaker wares.¹⁰¹ Besides confirming the accuracy of Bonsor’s original discoveries, this also emphasises the pioneering role he played in understanding the prehistory of the Scilly Isles.

In sum, there are a few points worth noting about Bonsor’s campaigns. The most evident one concerns his ability to conduct complex multi-disciplinary research deploying archaeological, philological, and historical methods. All these lines of enquiry contributed significantly to building a solid investigation, that evolved in the light of new data without losing coherence. Bonsor notably conducted two different enquiries at once, keeping track of the sources, data and methods needed to develop each of them accurately. Instead of bringing the campaigns to an end when he failed to identify a Phoenician presence in the Scilly Isles, Bonsor modified the objectives of his research. In doing so, he displayed a considerable degree of reflexivity, particularly given how deeply-rooted the hypothesis was in contemporary scholarship. In this context, the unexpected results, as well as the difficulties that Bonsor had expressing himself in English in an academic context (with French being his mother language), it is perhaps unsurprising that he never published the results of his campaigns in Scilly, despite the insistence of his contemporaries.

A second point to underline is his fieldwork expertise, which Bonsor demonstrated through the accuracy of his documentation of the archaeological record. As has been shown, this was of sufficient quality to inform later excavations on the same sites. Sadly, Bonsor’s explorations of the Scilly Isles remained almost unknown to most of academia until the second half of the twentieth century. It is hoped that this paper goes some way to restore his work from historiographic oblivion.