(Flint)Lock, Stock and Two Smoking Barrels: 18th 19th Century Gunflints from Dutch and British Archaeological Contexts

1

(Flint)Lock, Stock and

Two Smoking Barrels:

18th-19th Century Gunflints from Dutch and British Archaeological

Contexts

Anna Kohanoff

(Flint)Lock, Stock, and Two Smoking Barrels:

18

_-19

_{Century Gunflints from Dutch and British Archaeological Contexts}

Anna Kohanoff

Faculty of Archaeology University of Leiden

June 2019

Thesis MSc Material Cultures

Supervised by Prof. Annelou van Gijn Material Cultures

i

C

ONTENTS

Acknowledgements ... iii

1 Open Scene (Introduction) ... 1

1.1 Introduction ... 1

1.2 Materials and Methods ... 2

1.3 Aims ... 3

1.4 Research questions ... 3

1.5 Chapters ... 4

2 The EIC, the VOC, Shipwrecks, and a Quest for Fire-Starters ... 6

2.1 Gunflint Contexts in the 17th_-19th _{Centuries ... 6}

3 History of Gunflint Research ... 11

4 The Rolling Stones (Methodology) ... 16

4.1 Introduction ... 16

4.2 Materials... 16

4.2.1 Nepal (Kathmandu, Nepal) ... 16

4.2.2 The Rooswijk (Goodwin Sands, South Coast of England) ... 19

4.2.3 NB6 Wreck (Noordoostpolder, Province of Flevoland, Netherlands) ... 20

4.2.4 OL79 Wreck (Oostelijk, Province of Flevoland, Netherlands) ... 22

4.2.5 OH48 Wreck (Noordoostpolder, Province of Flevoland, Netherlands) ... 22

4.2.6 Experimental Material ... 23

4.2.7 Other Comparative Material... 25

4.3 Equipment ... 25 4.4 Provenance ... 28 4.5 Manufacture ... 29 4.6 Typology ... 31 4.7 Use ... 32 4.8 Discard/PDSM ... 34 4.9 Residues ... 35 4.10 Experimental Archaeology ... 35 4.11 Process ... 37

5 Bang Bang… (Analysis) ... 38

5.1 Terminology ... 38

ii

5.3 Rooswijk (Goodwin Sands, South Coast of England)... 50

5.4 NB6 Shipwreck (Noordoostpolder, Province of Flevoland, Netherlands) ... 51

5.5 OL79 Shipwreck (Oostelijk, Province of Flevoland, Netherlands) ... 58

5.6 OH48 Shipwreck (Noordoostpolder, Province of Flevoland, Netherlands) ... 59

5.7 The Experimental Gunflints ... 61

6 Viva the Gunflint (Discussion) ... 69

6.1 Provenance ... 69 6.2 Manufacture ... 71 6.3 Typology ... 72 6.4 Use ... 74 6.5 Residue ... 78 6.6 Discard/PDSM ... 80 6.7 Flintlock Artillery ... 82

7 Close Scene (Conclusion) ... 84

7.1 Flint, Morphology and Use-Wear Manifestation ... 84

7.2 Use-wear and Use-Life ... 84

7.3 High-power vs Low-power ... 85 7.4 Terminology ... 86 7.5 Final Comments ... 89 Bibliography ... 90 Figures ... 95 Tables ... 98

iii

A

CKNOWLEDGEMENTS

I would like to thank my supervisor, Professor Annelou van Gijn for guiding me through the study of lithics, for facilitating contact with Prof. Dr John Whittaker, and for all the time and effort she has put into helping me get this thesis cleaned up, in order and ship-shape. I would like to thank Dr Martijn Manders who has checked my work and got me involved with the Rooswijk material, helping me to gain access to the gunflints from this shipwreck; and as such also to thank the RCE for this access.

For that purpose, I would also like to thank Mark James for facilitating access to the laboratory at Fort Cumberland, to Nicole Schoute who helped me during the period of laboratory work, and to her again and Kim Roche for sending me updates and information throughout the year. Thanks to the whole MSDS Marine team for the warm welcome at the Rooswijk 1740 open day and meetings, and for their assistance throughout.

I wish to thank Mr Joran Smale at the Batavialand museum in Lelystad for helping me to select material and for facilitating access to the chosen gunflints and the museum itself for the loan of the gunflints. Thanks also to Dr Eileen Gregg for showing me the collection of

HMS Invincible gunflints at the Museum of the Royal Navy in Portsmouth.

Thanks to Professor Corrie Bakels and Dr Mike Field for helping me identify pollen when I was at a loss, and to Professor Patrick Degryse in identifying residue. Thanks also to Andrew Sorensen for talking me through fire-stones and facilitating flint striking experiments. I would like to thank Prof. Dr John Whittaker for the experimental gunflints, and for all the useful information he has provided me regarding his own research. I also would like to thank Mr Torben B. Ballin for providing me with an extensive reading list which has proved invaluable during the past few months.

Thank you also very much to Mr Eric Mulder and Mr Lou Jacobs for helping me in the material cultures laboratory and for the interesting conversations during long lab sessions. Finally, I wish to thank my family who have been incredibly supportive and tolerant throughout my academic studies and beyond, and for putting their time and energy into proofing this thesis.

1

This thesis explores the life of gunflints from Dutch cargo ship and mercantile contexts, and from a military Nepalese British context through a use-wear and contextual perspective. Typological assessment and microscopic analysis of micro-wear and residue are employed in tracing the origin, the use-wear, the use-life and death of gunflints from these archaeological contexts. Experimental material is employed for comparative analysis of the use-wear. The resulting conclusions focus on the relationship between the use-wear and the historical contexts of the samples; the adequacy of high-power analytical techniques for use-wear analysis of gunflints and the limitations imposed by current lack of standardisation across this field of study. Analysis by the high-power approach has shown that use-wear varies in manifestation and degree across flint type, whose state and hence study is impacted by depositional factors. Furthermore, a cross-referencing of past publications has highlighted key issues in the modern study of historical gunflints. Finally, a methodology is proposed for the analysis and data collection of distinctive morphological features.

1 O

PEN

S

CENE

(I

NTRODUCTION

)

1.1 I

The English-language term “flintlock” was coined by the year 1683 when it was first officially recorded – however, it is known that flints had been used in weapons for at least eighty years by that time, and also that the term “firelock gun” was already in use from the year 1547. Then, in the seventeenth century, flintlock guns were used in parallel with the earlier matchlock weapons, which had to be lit externally with tinder rather than within the cock of the gun itself like in the flintlock (Witthoft 1970, 15).

The need for systematically produced fire-starters – gunflints – arose with the need for a reliable method to ignite gunpowder in the advent of the first flintlock mechanised firearms, which debuted in their final standardized form in the 1620’s (Altamura 2013, 17; Witthoft 1970, 15). The invention of the flintlock mechanism was based on the ancient knowledge of fire starting by the use of striking flint to metal to ignite a spark, with an early written record being described by Pliny the Elder of a silex (flint) being struck against a

2

clevis (key). The technology of flint knapping had been a long-lost art until the invention of

this flintlock mechanism; with gunflint factories beginning as a cottage industry and developing into a fully-fledged operation by the peak of the gunflint producing period in the mid eighteenth century (Witthoft 1970, 20). Later, these factories declined as the flintlock fell into disuse due to new weapons being developed and the latter became more widely available. However, in countries on the fringes on the expanding European empires, flintlock firearms were a common weapon for a more extensive period, as the newer weapons were only made available once they were no longer the military standard and were decommissioned (Whittaker and Levin 2019, 3, 21).

Gunflint production followed two major streams of development – the English and the French methods. Hence, research has generally followed in one or the other direction. However, the cultural interactions of the time period in question beg for an analysis of these technologies in integration rather than in parallel. It therefore follows that the history of this research should be discussed chronologically as a whole, rather than in two - or indeed more, parts, thanks to the growth of international movement and trade over this time.

The aim of the study of gunflints, is to establish a provenance, to distinguish manufacture techniques and technologies, to analyse the use-wear on the gunflints, and to use these analyses alongside experimental pieces to gauge the approximate use life of an individual gunflint. The provenance will be approached using published gunflint literature and visual references; looking at the typology of the flint and manufacture technologies approached in the said references. The use-wear on archaeological and experimental pieces will be analysed and compared to assess the approximate amount of times a single gunflint was used to create a spark, and hence, how long one flint was useful for before being discarded.

1.2 M

M

The material that will be sampled comes from five different archaeological sites within three different contexts, and an experimental group. The assemblages are from an English East India Trading Company (EIC) context from Nepal, a Dutch East India Trading Company

3

(VOC) context from the Rooswijk shipwreck, and three from Dutch cargo shipwrecks – NB6, OH48 and OL79. The experimental material has kindly been supplied by Professor John Whittaker of Grinnell College, Grinnell, Iowa.

The methodology employed will encompass low-power techniques, coupled with a high-power approach, in the form of optical and reflected light microscopy respectively.

1.3 A

The aims of the thesis are to identify defining features of gunflint samples from EIC, VOC and Dutch cargo contexts. These features are related to morphology, flint type, coarseness and use-wear patterns; which will provide an insight into the fabrication and use of gunflints.

A high-power analytical approach will be applied, with the aim of gaining insights into its appropriateness for this type of research. The results will be used to assess to what extent it benefits the area of gunflint research with comparison to a low-power approach.

1.4 R

• What is the relationship between intrinsic gunflint properties and the use-wear it exhibits, including its period of usability?

• Is there a benefit to a high-power approach to use-wear analysis in gunflint research as opposed to a low-power one?

• What can use-wear, the depositional context and historical data tell us about the use-life and death of gunflints from military contexts in contrast to merchant and civilian contexts?

4

1.5 C

Chapter two will provide a general overview of the historical context of the gunflint samples. This refers particularly to the eighteenth through to nineteenth centuries in the respective geographical regions, with a particular reference to the English and Dutch East India Companies.

Chapter three will discuss previous research into gunflints, from its beginnings in the late nineteenth century right up to present day. This encompasses approaches such as experimental archaeology, use-wear analysis, and much more innovative approaches in the form of petrographic analyses through thin section analysis, LA-ICP-MS and CODA1_.

Chapter four gives a detailed description of the methodology that has been employed in the present paper. Here the samples are described, and the approaches that have been employed are recounted in detail with reference to the data that has been collected. The chapter is broken down into the following categories – materials, provenance, manufacture techniques, typology, use, and post depositional effects.

The fifth chapter gives the results of the analyses, broken down into categories based on the sample contexts, these being: Nepal, the Rooswijk shipwreck, the NB6 wreck, the OL79 wreck, the OH48 wreck, and the experimental material. General features of each collection will be noted, as well as the more distinctive features and observations made of specific gunflints within them.

Chapter six will then discuss the collected data. The information will be used to explore insights into the manufacturing technology of gunflints, their use, their deposition and to the appropriateness of high-power analytical techniques for this purpose. Further, it will

1 _{Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) and Compositional}

Data Analysis (CODA) are chemical analyses which give elemental, molecular and isotopic data which may be employed in determining raw material provenance.

5

aim at placing these insights into a discussion with regards to their historical contexts – and, hence, to the greater social role of these objects through the time periods in question.

The seventh and last chapter will make final remarks and conclude this paper, hopefully having provided a clear insight into the life and death of the specific gunflint artefacts being analysed. The chapter summarises the research and proposes future avenues for investigation.

6

2 T

HE

EIC,

THE

VOC,

S

HIPWRECKS

,

AND A

Q

UEST FOR

F

IRE

-S

TARTERS

2.1 G

C

17

-19

C

The beginning of the 17th _{Century marked the beginning of largescale European trade in}

the East Indies and into Asia and the Indian subcontinent. Small European companies had begun sailing there in the late 16th _{Century in the hopes of making their fortune in the spice}

trading industry. However, this abruptly changed at the turn of the following century as the Dutch and British restructured their trading organisations; and thus took up the mantle as the major intercontinental traders (Gaastra 2003, 13-17). Amongst the cargo of these vessels were gunflints in abundance that would end up in the East Indies, South Africa, the Americas, over Europe, and in some cases, submerged at the depths of the seas, lost en

route to these destinations (Schrire 2014, 222-8).

On the 31st _{of December 1600, the English East India Company (hereafter the EIC) was}

established, with the first monopoly grants being given by Queen Elizabeth I. Funding was initially based per voyage, with shares for its maiden voyage sold for capital to approximately two hundred investors. This was enough to pay for four large ships, a small supply ship, and close to five hundred men. Company servants came to take advantage of their position and began to partake in independent private trade, this internal corruption negatively impacted the EIC. This ended abruptly in 1787 under the newly appointed Governor Charles Cornwallis, who increased salaries to compensate for the loss of this private, if illegal, income (Webster, 2009, 24). The EIC came to have a minor power role on the South Asian coast in the Seventeenth Century, with its sphere of influence creeping increasingly inland into the 1750’s. The EIC would promptly develop a partly military focus, rather than a purely commercial one, as had been initially intended at the birth of the enterprise (Roy, 2018, 195; Webster, 2009, 18-23).

The Dutch East India Company (hereafter the VOC) was founded two years later - in 1602 and was hence run under the governance of the Heeren XVII. It comprised sixteen members distributed amongst six centres and a seventeenth member appointed on a rotational

7

basis. These members were each in possession of ships and storage houses. As the company grew, so did its’ complexity. This demanded the establishment of

buitencomptoiren (external offices) outside the Dutch Republic, and, thus also in the

proceeding years, of the Hoge Regering (High Government of the Indies) based in Batavia (modern Jakarta) – Asia (Sgourev and Van Lent, 2015, 936; Gaastra 2003, 66). The VOC became one of the defining features of the Dutch Golden Age, along with significant developments in the arts and cartography; in relation to these developments – ships becoming a common feature in paintings (Sutton, 2015, 1-2). The charter set out in 1602 allowed the VOC to make treaties with local Asian powers; and in the process, to hire soldiers and construct their own forts. This would be officially carried out on behalf of the States-General of the Dutch Republic, with the VOC being considered its executive tool. In reality, however, the regulations originally laid out in 1602 regarding the relationship of the governors of the VOC and the States-General, were promptly ignored, with the VOC acting more autonomously than originally supposed to. Within the first ten years of the establishment of the company, it became clear that it was not purely a commercial venture; but that it also served a previously unprecedented military purpose, which was unpopular with the initial shareholders from when the organization was originally conceived (Gaastra 2003, 23). The Governor-General and the Council in Asia would get involved in local warfare and treaty settlement, and the High Governor would also take initiative in Asian internal affairs despite the lack of support from the directors in the Dutch Republic (Gaastra 2003, 66). In 1665 the VOC gave economic aid in the form of a loan to the States-General in relation to the Second Anglo-Dutch War (1665-1667), solidifying its’ role in European military operations in the process (Gaastra 2003, 26).

From 1660, the competition between the EIC, the VOC and other, smaller groups of European traders intensified, in part as a consequence of independent changes in social and trade dynamics in Asia and Europe. As a result, the Eighteenth Century appears to show that there was an inverse correlation between the difficulties of the VOC and the increasing success of the EIC. The Third Anglo-Dutch War (1672-1674) left the VOC in financial difficulties – having to rely on the States of Holland and Zeeland for loans. A century later, a combination of bad book-keeping, corruption, inadequate financing and the Fourth Anglo-Dutch War (1780-4), left the VOC bankrupt and particularly vulnerable after a bout of Malaria in Batavia severely depleted its military manpower. By 1796 the

8

company had to be nationalized, until it finally ceased to exist altogether in 1800 (Sgourev and Van Lent 2015, 936-7; Gaastra 2003, 27, 164, 173). For the EIC, on the other hand, 1796 marked its peak, with eighty-four ships being sent out in just the one year; despite the change in its organization due to a shift to colonial rule following the Battle of Plassey in 1760, and also the subsequent financial difficulties and debts. Even with all its issues, the EIC continued to financially aid British generals in their battles in the Indian subcontinent right up to the turn into the next century. The year 1833 marked the end of the EIC’s commercial business as it finally lost its privileges to monopoly, thus surviving the VOC by a third of a century. Despite this drastic change, by 1840 the EIC controlled the majority of the Indian subcontinent, following a set of consecutive military successes that solidified the British colonial precedent (Webster 2009, 22, 24, 27,44; Roy 2018, 195).

As the EIC became more and more involved in the Indian subcontinent, so the resistance to this power increased. Furthermore, internal conflict between local rulers meant that the EIC would have to end up taking sides in military issues (Roy 2018, 195-6). Around 1815, the small kingdom of Nepal – situated to the north of India, decided to descend south and invade its neighbour. The EIC, which by this time had come to represent British interests and had begun to work alongside the British army, suffered multiple losses, until the Nepalese were finally defeated in 1816. Nepal would remain independent under the Treaty of Segauli, under the terms that the EIC would now have a monopoly on Nepalese trade. Furthermore, the EIC was given the liberty to recruit local soldiers, namely the Ghurkha regiment that had impressed greatly with their military skill (Whittaker and Levin 2019, 2).

The Nepalese gunflint assemblage originates from this British Imperialist context, having been discovered by the sack-load in an armament cache whose remnants consisted of decades-worth of decommissioned military weapons. Found alongside the flints were “Brown Bess” flintlock muskets – versions from both before and after the British-Nepalese conflicts, percussion muskets, cartridge rifles, Nepalese copies of the former, as well as small non-military guns, artillery, early machine guns and Enfield bolt-action rifles. The building housing all this old stock was bought in 2003 by Christian Cranmer who owns International Military Antiques (IMA), a stockist of discontinued and decommissioned militaria. The weapons used by the Nepalese were often old, obsolete weapons rejected by the British army – essentially an armament hand-me-down situation (Whittaker and

9

Levin 2019, 2-3, 21). The British allied themselves to multiple local powers in the Indian subcontinent, which provided men to the British regiments, but also often creating independent regiments with hybridised local and British features. The Nepalese manpower aiding the British came in the form of the Ghurkha regiment, which proved useful for suppressing the Indian rebellion that ensued in 1857. The Nepalese forces were armed with these older flintlock firearms from the early nineteenth century, until approximately 1860 (Whittaker and Levin 2019, 3; Roy 2018, 196).

During this period of extended trading networks, increasing warfare and of piracy, the need for artillery on board ships became more pronounced (Borschberg 2013, 47). Merchant ships such as the Dutch Rooswijk had approximately twenty-four cannon, as well as many pistols and muskets; the English warship HMS Invincible which sunk next to the former was equipped with seventy-four cannon amongst other weapons, and even the small Dutch cargo ship recorded as the NB6 wreck had a number of weapons – including a rapier and a musket. Thus, the weapons market boomed, with new technologies developing as seafaring became more dangerous, and international hostilities grew (Rediker 2004).

The flintlock mechanism was introduced into artillery at the turn of the Seventeenth Century so as to replace earlier, less efficient ignition systems (Altamura 2013, 17). The gunflint is clamped in the cock of the mechanism. As the trigger is pulled, it slams down and the edge of the gunflint strikes the steel frizzen on the other side. This action creates sparks, in the same manner as the manual flint and firesteel action, which subsequently ignites the black gunpowder. The force created by the combustion of the gunpowder propels the ammunition which is loaded into the weapon (Watt and Horowitz 2017, 1; Whittaker and Levin 2019, 5). This mechanism, which came to be used in all sorts of artillery, from pistols, muskets and even cannons, where it is known as a “gunlock”, would have to be adapted to the type of weapon, in turn impacting the size of the gunflints used (Biagi, Starnini, and Beltrame 2016, 4). The orientation of the gunflint in the mechanism is chosen so that the flattest, straightest and least used edge strikes the frizzen – this allows a greater surface area to maximise spark production (Whittaker and Levin 2019, 5).

10

When in use, a gunflint will be clamped in the cock of the gun, often with a patch of leather to support it. When asymmetrical in thickness, it is the thicker side which is placed toward the inside of the cock mechanism, with the acutely angled end facing out towards the frizzen. When the gunflint has multiple acute edges, the position may be adjusted, and the gunflint rotated along the horizontal axis to use multiple edges. In the United States, the gunflint is generally positioned with the ventral side up, dorsal side down. However, it is found that elsewhere the opposite also occurs – with the smaller dorsal surface facing up, and the larger ventral side facing down (Ballin 2012, 117).

Despite the advent of new, more modern artillery in the nineteenth century - which no longer used the flintlock mechanism, gunflints continued to be circulated. However, unlike the large caches from military contexts, the flints from this period became more prominently found in smaller quantities in civilian, or small trading contexts; perhaps due to the greater availability of older weapon types to these non-military individuals, or perhaps to be reused as fire-stones (Whittaker and Levin 2019, 16). This will be discussed in more depth in the discussion chapter of this thesis.

11

3 H

ISTORY OF

G

UNFLINT

R

ESEARCH

The study of gunflints has been sporadically carried out since the late nineteenth century, being led by individuals of various backgrounds – from weapons aficionados, geologists of sorts, and specifically archaeologists in more recent times. The research that has been carried out covers experimental archaeology, petrographic analysis and use wear analysis. This means that a great range of aspects of the gunflints has been investigated, and this data can then aid further study, with the aim of understanding technological aspects, and also the role of these artefacts in the economic and military spheres.

An academic interest in gunflint production began in 1879, with S. B. J. Skertchly’s publication within the Memoirs of the Geological Survey; whereby the methods by which gunflints are manufactured were discussed. His analysis began with a discussion regarding the geological positioning of the flint in the sedimentary layer and thereby its properties. The paper progresses onto the tools used for flint-knapping, followed by the method of extraction from the geology, and finally onto the techniques for the manufacture of the gunflints themselves. He continued to discuss the Brandon mines in England more specifically, and finally gave descriptions of the morphological features of different gunflint types (Skertchly 1879, 1-64).

In and around the same time, in the years preceding Skertchly’s publication, Sir John Evans (1868) – an English firearm enthusiast, and Sven Nilsson (1878) – a Swedish archaeologist, carried out the first technical experiments of flint knapping in the modern period. Personal experience with their own guns, and with knapping of gunflints for these, made them adequate candidates for making studies and descriptions of lithics, gunflints included. They would then go on to visit contemporary gunflint manufactories where they observed the practice of laminar production for tool manufacture (Altamura 2013, 19).

In 1937, decades following S. B. J. Skertchly’s publication, A. S. Barnes wrote about the gunflint industry of the English blade type. For his research he visited the Brandon mines in an attempt to collate the details of gunflint fabrication at this site; where he also attempted to knap a piece himself. The paper goes through the method of knapping

12

including tools and identifying which parts would be discarded or kept. Furthermore, Barnes went on to demonstrate that retouch related to the method of fabrication is the feature which distinguishes the English type from the French one (Barnes 1937, 328-335).

It is impossible to speak about gunflint research without addressing J. Witthofts 1966 seminal publication on the history of gunflints. Despite its’ Amero-centric theme, he made a useful chronology of the development of gunflints in Europe (and outside) which is useful for understanding manufacturing processes and advancements – thus aiding with establishing the provenance of manufacturing techniques if not of the flints themselves. Furthermore, he highlighted issues regarding this flint provenance, which will become more relevant when discussing the gunflint samples themselves (Witthoft 1970). In his discourse he also speaks of other individuals who had previously attempted to classify gunflints; namely the aforementioned A. S. Barnes, and L. Pfeiffer (Witthoft 1970, 43). However, some criticism of his work has ensued, focussed predominantly on the theme of flint provenance, and the unreliability of his techniques (Whittaker and Levin 2019, 2). Furthermore, recent advancements in petrographic analytical technologies bring into question Witthofts assessments and shed light on alternative interpretations (Whittaker and Levin 2019, 14).

In 1975, Stephen W. White wrote on the history of European gunspalls, and in so doing, actively argued against Witthofts attribution of a Dutch origin to the wedge-shaped form. In contrast, he attributed them to an English origin and dated them to approximately 1650-1770. This chronology was proposed based on the introduction of the blade-type gunflint at the end of the eighteenth century into England, a technology which White suggests was a French import rather than an independent local advancement in its own right. Furthermore, he states that there was a spall (or flake) type production centre in Denmark, and also that the finished product has been found concurrently at this time at American and French sites (White 1975, 65).

1984 saw the proposal of typo-morphological categories by De Lotbiniere in the Journal of

Nautical Archaeology and Underwater Exploration. These categories have consequently

13

gunflints into the four proposed categories. The focus of De Lotbiniere’s publication discusses gunflint shape, colour and size – in effect, as it is entitled; ‘gunflint recognition’ (Ballin 2014; de Lotbiniere 1984; Whittaker and Levin 2019; Austin 2011).

More recently, in 2011, Austin published a set of drawings in his report on Fort Brooke, Florida. These suggest similarities to the de Lotbiniere typology, however, the dimensional ratios differ somewhat, along with combinations of morphological features. It is important to note that in this case the classification is site-specific, with an attempt to use the former typo-morphological system. The names employed relate to the earlier publication, but there clearly were some complications in their application as there is some divergence from the former system (Austin 2011).

Figure 2 1) D-shaped gunspall; 2) square gunspall; 3) square blade gunflint with two dorsal arrises; 4) square blade gunflint with one dorsal arris. H= heel; E= leading edge; B= bulb of percussion; d= "demicone". (Ballin

2012, 119)

Figure 3 A) gunspall; B) round-heel blade; C) untrimmed blade; D) double-edged blade; E) square-heel blade. (Austin 2011, 87)

14

J. J. Durst has since the turn of the twenty-first century focussed his research on gunflints from VOC contexts in South Africa. According to him, the need for gunflints arose in the 1530’s with the invention of the snaphaunce from Germany, and the subsequent Spanish snaplock. He argues that the first flintlock mechanism originated in France, which is believed to have been developed there in the 1630’s. Durst’s publications discuss the raw materials, morphologies of gunflints, and also the scientific methodologies which may be involved in distinguishing the source of the flint. These analyses and the discussion of them is particularly useful for the present study, in terms of contextualising the gunflints within the greater sphere (Durst 2014, 221-229).

Amongst this contemporary gunflint research, T. B. Ballin is pre-eminent in his field, with an abundance of publications with focusing on varying archaeological contexts. He makes a clear distinction between the origins of gunflints by their morphologies, flint colour and fabrication technologies. Ballin places the flake based, round-heeled gunflints chronologically anterior to the others, being a common form across Europe. This type seems to have been kept in use in Scandinavia, whereas in the rest of Europe, the blade technology was later developed, adopted and adapted. The French developed their own gunflints sometime between the late seventeenth to early eighteenth centuries, with a broad body and a curved heel. The British gunflints, on the other hand, became elongated, and were given a straight heel. Ballin also associates the flint colour to approximate geographical regions. However, this is most relevant in discussion when matched to a morphological type (Ballin 2014, 45).

A recent development aims at identifying the provenance of the flint with analyses focussing of Scandinavian ballast from a Norwegian shipwreck (Brandl et al., 2018). The techniques of comparative analysis, geo-chemical analysis of trace elements by LA-ICP-MS (Laser Ablation Inductively Coupled Plasma Mass Spectrometry), stereo-microscopic analysis of microfossil inclusions, and Compositional Data Analysis (CODA) statistical analysis, were employed. The use of petrographic analysis on this ballast is a step further in the application of scientific methodologies in an aim to identify the source of the material. The premise is that it is possible to differentiate between geochemical compositions, which when analysed can give an indication of the depositional environment of the flint (Brandl et al. 2018, 1 & 11-24). This methodology could in theory be employed

15

in the future (although will not be able to be employed for this thesis) for determining the source of gunflints, provided that there is enough comparative material, and that there is not an issue with destroying some of the primary material. It does, however, have complications, as flint can be fairly homogenous at times and, furthermore, the comparative material may not be available or yet recognised (Brandl et al. 2018, 34).

Whittaker and Levin published the latest of papers in relation to a sample of gunflints from the Nepalese armoury at Kathmandu (Nepal) (Whittaker and Levin 2019). The paper focusses on identifying the provenance of the flint by analysing the morphological features and colouration of the gunflints. The aim of this is to gain insight into the origins and organization of historic Nepalese military technology and the European resources acquired, and in some instances imitated, by the Nepalese forces (Whittaker and Levin 2019, 1-2).

This former gunflint research is a valuable tool for present day researcher as it sets precedents and highlights areas that require further investigation. This thesis uses this research as a resource in combination with primary high-power analysis and historical data, so as to give specific gunflint assemblages from British and Dutch 18th _{and 19}th _Century

16

4 T

HE

R

OLLING

S

TONES

(M

ETHODOLOGY

)

4.1 I

Use-wear analysis is a form of study by which the function of a tool or object is interpreted by the observation of physical modifications on its edges and surface. This practice initiated at the turn of the twentieth century through macro-wear analysis, but only really began to take form with the ground-breaking work of Sergei Semenov in the 1930’s. His seminal work introduced new methodologies of practice - including the use of experiments as comparative material, which paved the way to use-wear analysis as we know it today (Gijn 2014, 166; Marreiros, Gibaja Bao and Bicho, 2015, 5 & 8). The following chapter will introduce the material and methodology which will be employed for the study of the gunflint samples.

4.2 M

The gunflint samples vary greatly in size; this being due to logistical issues concerning the acquisition of materials. The material being studied has been kindly supplied by different organisations, thus the quantity and nature of samples have been predetermined by what has been made available by these organisations (Gijn 1990, 9). The samples are as such: fifteen gunflints from Nepal, nine from the NB6 shipwreck, two from the OL79 wreck, another two from the OH48 wreck, three from the Rooswijk wreck, and finally, ten manufactured and used experimentally by Prof. Dr John Whittaker of Grinnell College, Iowa – in total forty-one pieces.

4.2.1 Nepal (Kathmandu, Nepal)

The Nepalese material comes from a cache in a disused building which served as a Nepalese armoury up towards the later part of the nineteenth century. After Christian Crammer purchased the building, he began to explore the different rooms, and as he descended into the basement, discovered that he was treading on thousands upon thousands of stones. When taken back up into the light, it was deduced that they were in fact gunflints. As they were collected, it was estimated that there were approximately 1.3 million; thus the 15 used in this study represent a token sample from this context. The discovery of this cache,

17

Figure 4 Gunflints from Nepal: a, b - square blade with two dorsal arrises; c, d – square gunspall; e, f – gunspall?; g, h – round-heel blade; i, j – round-heel blade; k, l – square heel blade; m, n -D-shaped gunspall; o, p – D-shaped gunspall.

its purchase, and the discovery of what lay within has been described by Cranmer himself in a documentary filmed in 2007, where he disclosed the manner in which the objects were found and treated after being found (Whittaker and Levin 2019, 2, 20; Cranmer 2004, 18, 40-2). b a c d e f g h i _{j k l} m _n o p

18

Figure 5 Gunflints from Nepal: a, b – D-shaped gunspall; c, d – D-shaped gunspall; e, f – square heel blade; g, h – square heel blade; i, j – square blade with two dorsal arrises; .k, l

– Square blade with two dorsal arrises; m, n – D-shaped gunspall; o, p – gunspall?.

a b c d

e _f g h

i j k _l

19

4.2.2 The Rooswijk (Goodwin Sands, South Coast of England)

The Rooswijk was a Dutch East Indiaman trading ship, built in 1737, sailing her maiden voyage in that year. Unfortunately, on the first day of her second voyage – from the Texel Roads to Jakarta, in January of 1740, she was struck by savage weather and sunk onto the Goodwin Sands – just off the south coast of England. Here, she joined other vessels taken down by the treacherous regional conditions (https://www.world-archaeology.com; https://wrecksite.eu). The presence of gunflints on this ship is unsurprising as multiple cannon and firearms were found associated with the wreck. Furthermore, the presence of items associated with illegal trade – in particular from Mexico, suggests that the ship could easily have been the target of pirates along her trading route (http://rooswijk.huygens.knaw.nl/).

The wreck site is owned by the Cultural Heritage Agency of the Netherlands (hereafter the RCE), it is currently protected and managed by Historic England (https://www.bbc.co.uk). A preliminary salvage excavation of the wreck was carried out in 2005 (http://rooswijk.huygens.knaw.nl). Since, there have been rescue excavations in two seasons in 2017 and 2018 – this time undertaken by the RCE in collaboration with Historic England and MSDS Marine (https://msdsmarine.com; https://www.world-archaeology.com). Three gunflints were recovered from the site during the 2017 and 2018 seasons, and one from the previous 2005 season – unfortunately the latter gunflint could

Figure 6 Rooswijk gunflints: a, d - fragment; b, e – round-heel blade; c, f – round-heel blade. (Provided by MSDS Marine).

a b c

20

not be included in this study. Twenty-four cannons and a multitude of muskets were found in the wreckage, therefore there is reason to believe that there are more gunflints still at the depths of the sea, somewhere on the Goodwin Sands (http://rooswijk.huygens.knaw.nl).

4.2.3 NB6 Wreck (Noordoostpolder, Province of Flevoland, Netherlands)

The NB6 wreck is a late eighteenth century of a cargo ship of the tjalk type, discovered in a field in the Province of Flevoland in the Netherlands. The contents of the ship indicate that the port city of origin prior to its’ demise was Groningen (http://cultureelerfgoed.adlibsoft.com; http://www.verganeschepen.nl). The remains of the shipwreck were discovered during work in the Noordoostpolder in 1952 and were explored and subsequently excavated in 1955. The ship was dated based on the contents and this was established as 1787. Furthermore, tree ring dating set the date of tree felling to 1769, meaning the ship was in use no longer than eighteen years after. The cargo of the ship consisted of tiles, but was also carrying soldiers from Muiden to Hoorn at the time she sank (Manders and Kuijper 2015, 154; http://www.verganeschepen.nl). Nine samples are from the NB6 wreck, and a small sample of gunpowder from the same wreck was also kindly supplied by the Batavialand museum in Lelystad (Netherlands).

Figure 7 NB6 gunflints: a, b – round-heel blade; c, d – round-heel blade; e, f – round-heel blade; g, h – round-heel blade; i, j – round-heel blade;k, l – “other”.

a b c d

e f g h

21

Figure 8 NB6 gunflint: a, b – square blade; c, d – “other”; e, f – square blade.

a b c d

22

4.2.4 OL79 Wreck (Oostelijk, Province of Flevoland, Netherlands)

The OL79 shipwreck is of an eighteenth-century Dutch cargo ship, with the skipper being possibly identified as Pieter Jacobszoon Pondman (Velde, 2014, 6). It sank in the Zuiderzee in the Netherlands in the year 1796 and was discovered when the Flevopolder was created (Filatova and Van Popta, 2014, 99). A survey of the site was carried out in 1959, but it was only in 2013 that the ship remains were excavated (Manders and Kuijper 2015, 154). This was undertaken as part of a collaborative project initiated by the Flevoland provincial authority and integrating the Lelystad council, the University of Groningen, Nieuw Land and the Cultural Heritage Agency of the Netherlands. The project was initiated in 2007 titled International Field School for Maritime Archaeology Flevoland (IFMAF) as an educational as well as practical archaeological tool (https://maritime-heritage.com). Two flints were found at this site, and like the NB6 are curated by the Batavialand museum.

4.2.5 OH48 Wreck (Noordoostpolder, Province of Flevoland, Netherlands)

The OH48 shipwreck is that of a carvel built nineteenth-century Dutch cargo ship going by the name of the Lutina. She was excavated in 1976 after being discovered on the Beverweg – close to Swifterbant in the Noordoostpolder. The cargo she carried included shells and bricks which were being transported from Brielle to the Zwartzluis lime kilns by the skipper Jan Roelofszoon Kisjes and Reinder Tulp – his servant. The ship sank on the 20th of November 1888, being struck on that fateful night by a thunderous storm. The gunflints being investigated were found on board amongst other objects, including, the most striking

Figure 9 OL79 gunflints: a, b – “other”; c, d – round-heel blade.

a b

23

find: the skeleton of the skipper. (https://www.flevolanderfgoed.nl; https://www.omroepflevoland.nl). This final set of samples from the Batavialand museum consists of two pieces.

4.2.6 Experimental Material

The experimental material is supplied by Professor John Whittaker, of which the morphological choices are modelled on the Nepalese samples. However, the flint used in the experiments is of North American origin as opposed to the European and Nepalese flint used for the pieces in the Nepalese sample. The gunflints are based on blade blanks, which however, have different finished morphologies. Whittaker noted in his experiments what edge was used, as well as whether the dorsal surface faced up or down. The number of shots fired with each gunflint was noted – thus providing adequate information for comparison, albeit being most useful for material of a similar coarseness of grain.

Figure 10 OH48 gunflints: a, b - gunspall; c, d – square blade with two dorsal arrises.

a b

24

Figure 11 (a – t) – Experimental gunflints based on gunflints from the Nepalese armoury in Kathmandu; u, v – leather patch.

a b c _d e _f g h i j k l m n o p q r s t u v u u u _u u u u u u u u u u _u u u u u u u

25 4.2.7 Other Comparative Material

A visit to the Museum of the Royal Navy in Portsmouth (England) has provided a collection of comparative material to be used comparatively against the archaeological samples – in particular the Nepalese sample, for this present study. This material is from the HMS

Invincible, which is notable for its English context and its’ relationship to the EIC. The ship

was originally a French 74-gun warship named L’Invincible built in 1741 and launched on the 21st _{of October 1744. She was captured by the English a few years later on the 3}rd _of

May 1747, becoming a model for future English ship building. Ten years later in 1757 she got caught in a hurricane which resulted in her sinking onto a sand bank off the South coast of England, not far from the recently wrecked Rooswijk. The museum has 217 gunflints, which is representative amongst the thousands which were recovered from the wreck – the most notable aspect of which being the uniformity of the flint material, as well as the standardisation in object morphology (https//:historicengland.org.uk).

A second source for comparison is used, particularly for analysing gunflints from amongst the archaeological samples which display signs of having changed purpose into fire-stones. This material consists of three pieces of flint varying in grain coarseness and size, which were consequently struck by the author with a firesteel repeatedly on one location of the stone. The use-wear typical of this action is of large edge removals concentrated on one part of the gunflint, which with repeated use results in a crescent dent (D. S. Buscaglia, Humanas, and Aires 2016, 11).

4.3 E

The techniques presently employed for use-wear analysis include the use of low and high-power microscopy – in the form of stereo and reflected light microscopy; these being used in tandem with experimental archaeology (Evans and Donahue 2005, 223-4; Gijn 2014, 166; Marreiros, Gibaja Bao, and Bicho 2015, 9). Analysis of retouch is the primary analytical approach which will be applied to the gunflints; this can be used to classify the retouch into use, intentional, and unintentional so as to see how the gunflints were used and if they were adapted. This can generally be distinguished by the location, position, distribution pattern and form of retouch. This can be achieved initially by observation with the naked eye in the aim to identify points of interest, and then these areas will be studied more

26

thoroughly with a Nikon SMZ800 stereomicroscope. Zones with particularly interesting features will be studied even more closely with a Leica DM1750 M reflected light microscope. Using microscopes with different magnifications and qualities is important to get a greater overall view of the gunflints, however for this purpose a microscope with such a high magnification and resolution such as Scanning Electron Microscopy (SEM) or laser confocal microscopy is unwarranted, and thus will be limited to the reflected light microscope (Marreiros, Gibaja Bao, and Bicho 2015, 41-2).

The Nikon SMZ800 stereomicroscope being used can give a magnification between 10x and 80x (without added accessories), has a zooming range of 1-8x and a working distance of 78mm (Catalogue Nikon Corporation 2017, 2 & 29). The low-power approach to lithic analysis, is limited in that it is only sufficient for initial observations – such as edge angles, edge damage, and fractures. It does not have the resolution to identify other traces which can give more insight into the nature of this damage and fractures (Marreiros, Gibaja Bao, and Bicho 2015, 9). It is, however, sufficient to identify traces of wear and residue, and to give a crucial general overview, which can then be examined more closely by higher magnifications for affirmation (Whittaker and Levin 2019, 11).

The Leica DM1750 M reflected light microscope being used has a magnification range of x100-400. It has a variety of illumination options – brightfield, oblique and polarized light. This allows different aspects of the objects to be studied as it can be adjusted accordingly to how the material interacts with the light type at different source angles. All the reflected light illumination is done with integrated Power-LED segmented lighting; which aids analysis when coupled with the different angles of illumination. The combination of features of the microscope makes it particularly suitable to investigate microwear and for topographical features. Furthermore, different accessories may be used with the microscope to obtain the best possible image (Catalogue Leica Microsystems 2015, 7; Marreiros, Gibaja Bao, and Bicho 2015, 9).

The Rooswijk samples were studied at Fort Cumberland (Portsmouth, England), which has a different microscope. The microscope employed is a Leica Wild M3B on magnifications

27

of x16 and x40, with eyepieces of x20. This stereomicroscope uses an external light source, in this case a Schott KL 1500 LCD which has a somewhat yellow hued light.

The use-wear that is examined is as follows: • Edge removals

• Residue

• Post-depositional surface modifications including patination • Striations

• Polish

As opposed to the stereomicroscope, a reflected light microscope can give a detailed view of the surface of lithic material. The nature of the fractures and polish have a different texture which is visible by this method due to the higher magnification, and manner by which the light interacts. This means that in theory, in combination with experimental reference material, the type of use-wear can be categorized and attributed to a specific action and material (Marreiros, Gibaja Bao, and Bicho 2015, 9). Furthermore, small particles, such as metal residues from the frizzen used in flintlock mechanisms, are more likely to be seen by this type of microscope (Whittaker and Levin 2019, 10-1). However, an issue with this type of microscope is that it has a restricted focus plane which makes sampling more complicated (Gijn 2014, 168). The Leica DM1750 M has the benefit of being able to be fitted to take photographs – in this case being used with the Helicon focus program; these photographs can then be layered to give a focussed image of the whole surface topography, as opposed to just a slice of it (Catalogue Leica Microsystems 2015, 11-12). However, due to the material, and more in particular morphological, nature of the sample material, it is often difficult to obtain an accurate representation of the surface.

The biography of the use-life and death of the gunflints is read through the interpretation of the typo-morphological features and use-wear traces on the gunflint surface. The provenance is examined through the knapping technology, object form and flint type; the life – including functional changes, through the interpretation of edge removals, crushing,

28

flaking and residues; and the death through the depositional context and post-depositional surface modifications.

4.4 P

Flint comes in a variety of colours and qualities; even within one location there can be huge variations in the nature of the flint quarried – related primarily to the depth it is from as explained by the sedimentation process of flint (Durst 2009, 24; Skertchly 1879, 5-8). Therefore, using flint qualities and chemical properties as a method of source identification is almost impossible - although the honey-coloured type is nearly always identified as French (Ballin 2014, 45; Durst 2009; Whittaker and Levin 2019, 7; Witthoft 1970, 39). Hence, the method employed to interpret the provenance of the gunflints will be by the analysis of their morphological features, measurement of dimensions, measurement of the front and rear edge-angles, and identification of general form; and thereafter categorizing them into one of the previously established and recognised (Ballin 2014, 51-6). This data can be used in conjunction with colour and grain coarseness to give an indication of the most probable provenance of the individual gunflint; when there is a relevant reference collection. The pioneering work of modern research by John Witthoft, and the publications by Prof. Dr John Whittaker and Torben Bjarke Ballin will be considered in this procedure (Witthoft 1970; Whittaker 2001; Whittaker and Levin 2019; Ballin 2014; Ballin 2012).

European gunflints from a great variety of contexts, are generally attributed to French and British Industries, although Witthoft had claimed that there was a Dutch workshop (due to flint colour and the existence of a stock factory) (Witthoft 1970, 34). However, this has been under question due to lack of substantial evidence, and appears to since have been disproven by the lack of a good flint source in the Netherlands (Durst 2013, 20; Ballin 2012, 118-9). Other European gunflint industries can be sourced to Italy, Austria, Germany and into Eastern Europe, however, these were minor industries in comparison to the French and British, at least until the later period of gunflint use (Whittaker and Levin 2019, 4; Zeebroek et al. 2010, 267; Ballin 2013, 1-4). French flint has multiple sources, with the Meusne region in North East France appearing to be the most prolific source for gunflint material. Witthoft explained that French flint had also been transported to Britain at the turn of the nineteenth century, this must be considered as it can explain the presence of

29

British style gunflints produced with French material (Witthoft 1970, 36, 41, 43). English gunflints, whether used at home or abroad, are produced with flint native to England due to the prevalence of high-quality stone in the British Isles – most notably at Brandon, but also in other regions. Ballin attributes different physical qualities in the flint to a source, thus narrowing down the possible locations of origin (Ballin 2012, 121; Karklins 1984, 51-2; Whittaker 2001, 382). Italian flint from Verona became prolific during Napoleonic occupation of the region, therefore adding another typology into the mix (Chelidonio and Woodall 2015, 1).

Gunflint assemblages outside Europe, however, may derive from a different source. Only some of the gunflints are imported from Europe, whereas a large proportion are actually believed to be of local stone and production. This assumption is based on the flint typology and the different skill level in the knapping of gunflints (Witthoft 1970, 52; Whittaker and Levin 2019, 7-8, 11-2). A step in attempting to solve these issues, is an experiment which was executed in an attempt to sort the flint by atomic absorption spectrometry. By this method, it has shown that it is possible to establish a probable generic European origin, albeit a non-specific one (Whittaker and Levin 2019, 8). With that in consideration, Prof. Dr Whittaker’s analysis of gunflints from a nineteenth century British stockpile in Nepal – which will be re-examined here, has suggested that the assemblage is composed of British type, some French type, and some “other” that do not fit either of the former typologies. The latter have been suggested to be a local production, however, as they are not yet formally recorded, they do not currently belong in a reference collection – thus they are for the time being, categorised simply as “other” (Whittaker and Levin 2019, 1).

4.5 M

A macroscopic analysis of the gunflints is sufficient to establish the technology employed in the manufacturing process. It has been generally established that there are two main gunflint manufacturing technologies - the flake technique, and the later snap-blade; the latter being considered to be more efficient as it produces less waste (Kenmotsu 1990, 98-9). The techniques can be identified by examining the gunflint morphology, and in so doing, by also identifying and studying the bulbs of percussion (Ballin 2012, 117-8; Barnes 1937, 330).

30

Gunflints may be manufactured by use of either flake (fig. 12. Pg 31) or blade blanks (fig. 13. Pg 31) which are obtained from prepared cores by having had the external cortex removed. Flake blanks are often struck off the ventral surface of the flint flake, which are then heavily retouched on all four edges, and occasionally also on the surfaces. Blanks from blades, however, are taken by the segmentation of blades struck from the flint core. The form of a gunflint is largely determined by the manner in which the blank is segmented, and the way in which it is retouched to give it its form (Arcos, Fernandez, and Rodriguez 1996, 111-3; Ballin 2012, 118; Whittaker and Levin 2019, 5-6).

In identifying the blade technique, useful morphological features on the gunflint include: a side snap fracture, the presence of a dorsal ridge, and of erailluer flakes (Kenmotsu 1990, 100). The Italian gunflints have been described as a “platform type” which is one derived from the blade technology, no flake types have been found amongst the Italian examples, which aids in classification of unknown samples (Chelidonio and Woodall 2015, 2-3). Gunflints that do not fit to either technology type, on the other hand, - for example some of the Nepalese ones, cannot be put into a clear typology, but are fairly uniform and are believed to be produced locally (Whittaker and Levin 2019, 7-8).

The morphology of a gunflint is not only dependant on the technology employed in manufacture, but also what part of the process it is derived from. What is meant with this, is that if a gunflint is produced using blade technology, the portion of the blade the individual flint is acquired from will impact its shape, size and potentially the number of dorsal arises it has; furthermore, in French and English contexts, these examples are often much retouched – to be almost straight, dorsal and ventral edges (Kenmotsu 1990, 2013, 98-100). A flake gunflint however, unlike the blade type, displays a different general shape type being less standardised and clean cut, often with less defined edges. Where the blade type is quite thick and has generally quite a standard thickness (apart from the rear and leading edges, the flake type is more of a wedge in terms of thickness, with one end being thinner than the other, in some contexts being more retouched than in others (Ballin 2012, 118; Kenmotsu 1990, 98).

31

4.6 T

Gunflints are generally easily distinguished from the standard tinderbox flint staple due to their clear morphological differences. However, something that should be kept in consideration is that a disused gunflint – no longer being useful in a firearm, was likely often recycled to be used with a fire steel; which was still a common tool during the period of gunflint use (Whittaker and Levin 2019, 16).

There are two methods for blank production; that is the blade technique and the flake method. The technology employed for this can be identified by the gunflint morphology – the location and orientation of the bulb of percussion, the gunflint shape, and the geometry of the dorsal side of the gunflint (Ballin 2012, 117-8; Barnes 1937, 330).

Figure 12 Figure 12 Reduction of a 'flaked flake': 1. platform remnant with bulb of percussion; 2. Ventral face of the 'flaked flake'; 3. decortication

flakes; 4. gunflint blanks. (Ballin 2012, 132)

32 TYPOLOGIES: Flake based • Gunspall • D-shaped gunspall Blade based

• Square blade gunflint • Rectangular blade • Square heel blade • D-shaped blade • Round-heel blade

When classifying into a typo-morphological group, we must be weary and keep in consideration that there could be a plethora of reasons for similarities in gunflints discovered in different geographical regions, and vice versa – differences in those from the same region. These can be for example, the hardness, brittleness or coarseness of the flint, as well as the colour. These physical features can vary greatly not only geographically, but also even within a single source site – in relation to the depth, and therefore the geological layer it is from (Skertchly 1879, 5-8; Whittaker and Levin 2019, 7). The typo-morphological features associated with each gunflint may in turn vary according to the skill of the flint knapper, and due to manufacturing mishaps that can happen even to the most proficient of craftsmen (Luedtke 1998, 36-7).

4.7 U

The examination of the use of the gunflints will be approached by macro and microscopic analysis in an aim to observe the manner in which use-retouch manifests, which edges were used to strike the frizzen, and also to an extent which part was clamped into the cock of the flintlock mechanism. The listed variables impact the form in which the following use-wear develops on a gunflint:

33 Variables:

• Morphology • Flint coarseness • Angle of use edge Use-wear:

• Edge removals – step, hinge and feather terminations • Erailleur flakes • Polish • Residue • Crushing • Striations • Patination

It is important to bear in mind the intrinsic properties of the flint type when carrying out use-wear analysis, as the grain size of the flint will impact the manner and degree in which use wear traces develop. For example, it is logical to expect a finer grained flint to display more intensive use than a coarser flint. Furthermore, the typo-morphological qualities of a gunflint, in particular with regards to the edge angles, may have an impact on the creation of edge removals, and the degree of their manifestation (Ballin 2014).

Gunflints and fire-starters can have essentially the same (although not exclusively) morphology; therefore, microwear analysis is a useful tool which may be used so as to distinguish between the two. This is relevant to this study as, for example, in the seventeenth century, both flintlock firearms and fire-starters were used simultaneously (Witthoft 1970, 37). Another possibility is also that these flints were multipurpose, however, these hypotheses require further testing through experimentation and comparison of wear marks.

34

4.8 D

/PDSM

The discard of gunflints generally results in the development of post-depositional surface modifications. This is primarily due to the disposal context of these objects, which means that once used to their maximum potential, the gunflints are collected in large containers where they hit each other, or in caches where they are easily trampled (Whittaker and Levin 2019, 3). Another possible source of post-depositional surface modification is from an excavation or extraction process. The identification of post-depositional surface modifications is approached by both a macro and microscopic approach, though the latter may be more effective in yielding results.

Gunflints are often recovered from maritime contexts – generally in the form of shipwrecks, be it sunken, or, as is common in the Netherlands, from terrestrial sites such as buried in a polder (Manders and Kuijper 2015, 153-5). Other gunflints have been recovered from battlefield contexts – in Europe for example, a large quantity has been found in Spain and France; and in Asia they can be also be found in armament caches (Witthoft 1970, 50; Whittaker and Levin 2019, 1-4). The effect of water – which can result from water-logged soils or indeed direct water on flint contact, can be observed in the form of a white patina – this being particularly relevant to gunflints recovered from a marine context (Ballin 2014, 45; Burroni et al. 2002, 1281). The white patina is generally attributed to alkaline solutions, which is therefore easily explained as sea water is alkaline due to the ions dissolved in the solution (https://www.lenntech.com). Furthermore, the wear on these flints can differ from those retrieved from terrestrial sites as they may be cleaner due to the cleansing action of the water. Furthermore, extraction from concretions, if done incorrectly or without due care, can result in damage of the gunflints in question. Gunflints from terrestrial shipwrecks however, such as the NB6 wreck, can display not only the effects of water, but also are more prone to post depositional surface modifications – due to an increased risk of trampling or damage during the excavation process which does not happen during an underwater excavation (Burroni et al. 2002, 1277).

Post-depositional surface modifications (PDSM) can make interpretation difficult as they alter the surface and polishes on the objects, thus complicating the microscopic analysis (Levi Sala 1986, 229; Burroni et al. 2002, 1277-8). In terms of gunflints, the action of

35

disposal or deposition into a cache – where this would not be done with care, could result in surface modifications unrelated to use in the flintlock mechanism. Furthermore, there is a possibility that the extraction of the gunflints from concretions – which is how a great proportion of maritime material is found, could also result in surface modifications. The issue is distinguishing between natural depositional modifications from use-wear as the effect can appear similar. Finding multiple alterations can help identify a process, and thus the real use surface and wear patterns (Burroni et al. 2002, 1278-9).

4.9 R

A further application of microscopic analysis is the identification of residues on the gunflint surfaces (Gijn 1990, 8). In this case, the residue of interest clearly is gunpowder, keeping a keen eye on how it is manifest and its distribution on the gunflint surface. It is important however, to also be attentive to any other possible residues which may give some other information, perhaps relating to the deposition context or other outlying factors. The questions of interest in this regard concern the distribution pattern and intensity of said residue, and the degree of preservation of gunpowder in relation to the archaeological context of the gunflints. The naked eye can be enough to establish the presence of residue – assuming that there is a substantial amount of it. However, the use of a stereomicroscope can be beneficial to identify if there are residues in small quantities. Reflected light microscopy, can aid in discriminating between residue types; and can therefore be used to establish the presence of gunpowder more specifically – when in doubt. Residue from the gunflints – both from historical and experimental, samples will also be compared to a sample of gunpowder recovered from the NB6 shipwreck with the aim of observing if there is a standard form of black gunpowder over these periods and geographical regions, during the period of flintlock weapon use, however not too much is expected from this aspect of the study.

4.10 E

A

Experimental archaeology allows us to directly compare the different actions taken on an experimental object, against an archaeological object for which these are an unknown. It is possible for us deduce how an object will look after use based on its morphology,