The Secret of Early Maps

The Secret of Early Maps
Talks / Lecture

Recording date: 
12 September 2016

Hear from Chet Van Duzer as he reveals secrets uncovered by multispectral imaging of the Henricus Martellus World Map, thought to be one of the most important of the 15th century. Dr Martin Woods, Curator of Maps at the National Library and Denyl Cloughley, Manager of Preservation Services, lift the veil on the history and ongoing challenges of preserving the Blaeu wall map of New Holland.


Transcript of ‘The Secret of Early Maps
Speakers: Stuart Baines (SB), Martin Woods (MW), Denyl Cloughley (DC), Chet Van Duzer (CVD)
Location: National Library of Australia
Date: 12/09/2016

SB:       Distinguished guests, ladies and gentlemen, welcome to the National Library of Australia. My name’s Stuart Baines and I’m the Library’s Assistant Director of Community Outreach. As we begin I’d like to acknowledge the traditional owners of this land, I thank their elders past and present for caring for this land we are now privileged to call home. 

It’s wonderful to see so many of you here tonight, a packed house which is always great. We’re lucky to be joined by map and preservation experts, Chet Van Duzer, Martin Woods and Denyl Cloughley. Chet Van Duzer is a board member of The Lazarus Project at the University of Mississippi which brings multispectral imaging to cultural institutions around the world. He has published extensively on medieval and renaissance maps and journals such as Imago Mundi, Terra Incognitae, Word and Image and Viator. He’s the author of Johann Schöner’s Globe of 1515, Transcription and Study, the first detailed analysis of one of the earliest surviving terrestrial globes that includes the new world. And with John Hessler, Seeing the World Anew, the Radical Vision of Martin Waldseemüller, 1507 and 1516 World Maps

His book, Sea Monsters on Medieval Renaissance Maps was published in 2013 by the British Library and in 2014 the Library of Congress published a study of Christopher Columbus’ Book of Privileges which he co-authored with John Hessler and Daniel De Simone. His book, The World for a King: Pierre Desceliers’ Map of 1550 is recently out from the British Library, and Brill just published a book he co-authored with Ilya Dines, Apocalyptic Cartography, Thematic Maps and the End of the World in the 15th Century Manuscript. Hopefully that’s all of the bad pronunciation done. His current book project is a study of Henricus Martellus’s world map, 1491 at Yale University based on multispectral imagery. 

Martin has been the Curator of Maps at the National Library of Australia since 2005. Originally from South Australia Martin is also President of the Australian and New Zealand Map Society. Martin was co-curator and Consultant Editor of the Blockbuster Maps Exhibition, Mapping our World, Terra Incognita to Australia which brought Australian and world cartography treasures to Canberra in 2013 to 2014 here at the National Library. He is the author of Where Are Our Boys? How Newsmaps Won the Great War, published last month. 

Since graduating from University of Canberra in 2001 Denyl Cloughley has worked as an objects conservator in museums, libraries, art galleries and private conservation companies throughout Australia and New Zealand as well as the United States of America. She currently manages the preservation team at the National Library of Australia. 

Tonight our speakers reveal the speakers behind some of the most important maps of the 15th century and lift the veil on the history and ongoing challenges of their preservation. Please join with me in welcoming Chet Van Duzer, Martin Woods and Denyl Cloughley. 


MW:    I'm only one of those three as you can see. This slide ... we presented a similar talk recently at a maps conference in Wollongong and it was great to have Chet there as well and he did a different talk so this is just one of his many talents, the talk tonight. I can tell you you're in for a really good show. 

What I want to do though is to begin with a talk about the Blaeu Map. The Blaeu Map was acquired during or just before we launched and presented the Mapping Our World conference and it was really just a series of circumstances that brought it to us. As you can see by that description it’s a very nondescript description. It says almost nothing about it and this was advertised in a ... auction house, Stockholm, and these details were provided, the price was modest and there was a stir and a storm of interest in it once people realised what it was. It was really one of the surviving ... the dinosaurs of the map world which is the wall map. The wall map as you could imagine is a very different beast to a little map like something of this size. An example wall map is out the front here for you to have a look at and a reduced ... a miniature atlas of a wall map is just there as well. These things were not meant to survive, they were meant to be shown, to be hung on walls as large as these here and through time and stresses and strains they simply are rolled up and thrown away and this map that we acquired in 2013 was found in a storage facility in ... after 50 years of residence and that’s the only provenance that we've actually got for that map. We know it was produced in 1663 off a 1659 copperplate or in fact 30 odd copperplates and that’s about all we know of its life. 

But I want to talk a little bit about what’s on the map. Now this is not this map, this is another map by Joan Blaeu, it’s his world map and this world map was produced in 1648 and if you know your Australian history you’ll know that’s quite close to the year Tasman completed his second voyage, 16 ... just four years previously Tasman had been out here on the second voyage and had mapped much of the north coast of Australia. 

As the Du ... it goes, the story of New Holland, people accept that the Dutch did not have a happy time, they lost ships, there was no commerce however they left us with the name New Holland. Before New Holland however they’d already produced new lands, Edel’s Land, Knight’s Land, Eendracht, Land of the Eendracht and Carpentaria and other names that have been added to the map. The Dutch actually added about 200 names to the map of New Holland as it became and about 25% of these still remain. However as I said they did not have a happy time of it and what’s behind the New Holland is really what I’d like to address. Denyl, for her part, will look at what’s behind the map itself, physically. 

So Tasman we are familiar with, Van Diemen, we know his land. Tasman is a bit of a disappointing figure in some respects. He was sent out by Van Diemen on no less than two occasions to map the south land so the story of the south land is essentially a mythical story but people suspected there was something there. The Dutch had been since the early 1600s finding traces of it around the west coast and the northern coast of Australia, the southern coast of Australia and Tasman was sent via certain prescribed routes to map the rest of it or at least limit the south land further. 

He succeeded in some respects but he missed the east coast of Australia, he mapped some of southern Tasmania, he mapped as well half of New Zealand but he missed the Cook Strait for example so he didn’t understand that New Zealand was two islands. On his second voyage he mapped most of northern Australia but missed Torres Strait so again Van Diemen’s ambitions ... and Van Diemen was an ambitious man, if he set his gaze on you beware and Tasman was the lucky soul to be picked by Van Diemen. Van Diemen became Governor General of the Dutch East India Company in 1636 and he set about quickly annexing Solon and extending Dutch trade through to Formosa, down to Malaya and he had some unfinished business with New Holland. New Holland was a place he had visited 20 years earlier on his way out to the East Indies in 1618, the ship, Mauritius, bumped into the west coast and moved on. 

This must have left some sort of an impression on Van Diemen and it fitted with his expansionary feel, his style to expand and continue to expand so he used the myth of the south land to fund these voyages hoping, taking a risk, a gamble and you’ve got to bear in mind that Van Diemen was an entrepreneur in an earlier life, a failed entrepreneur but being appointed Governor General of the Dutch East India Company was the right ticket for him. Two failed voyages. His reign as Governor General ended immediately afterwards and in fact he died in 1645 just a year after Tasman returned. 

But he’s not the only character and by now New Holland was in the hands of another entrepreneur. This is the subject of the talk Denyl and I are giving, is Joan Blaeu and Blaeu was part of the dynasty of mapmakers, his father, Willem Blaeu, had really set the cartographic standards that others followed and there was much competition between Dutch mapmaking families. 

Joan Blaeu expanded the company massively and he introduced the world’s largest atlas, the Atlas Maior, which essentially blitzed his rivals, it went to 12 volumes and thousands of ... literally hundreds of maps in that atlas and many thousands of pages. So if you were a well to do burgher in Amsterdam you would acquire the Atlas Maior and Blaeu was the man who introduced that concept. Bigger was better in the Dutch world of the mid-17th century. 

So ... but Blaeu is also known to have accumulated drawings, paintings etc. This was a methodology that he used at the expense of his rivals. He would get in at the last ... at the last possible moment and acquire ... if any of you have been involved in auctions etc you will recognise this behaviour. Someone will get in there at the last moment and they will put in a bid that stuns everyone else and walk away with the lollies. And Blaeu was very good at doing this and he wouldn't necessarily then publish it, he would just sit on it and he was chased by many creators, creative people who were doing their work. Blaeu’s style was to acquire these and when it suited him he may publish but it made him very successful. 

So when Tasman returned of course Blaeu was going to be the first person to publish a view of New Holland. Why New Holland? Does anyone know? Is it New Holland? And is it the only New Holland? Something to understand is this was the New Netherlands, the Dutch republic of the 17th century that had forged an independent status from Spanish Hapsburg rule and so it would be more logical to call it New Belgium ... sorry, Belgica which was the Roman name for some part of the Dutch provinces. But anyway by now Holland had become the dominant province, Blaeu himself came from Holland so while there was New Netherlands in ... there was at two New Hollands and New Netherlands in North America, already viable colonies by now, he chose New Holland and we’re still really not quite sure although Tasman applied New Holland to the map for the first time. 

So you see these traces of New Holland on the floor of the ... what is now the Royal Palace of Amsterdam which was the Dutch ... the town hall and if you’ve ever had the chance it’s a massive installation and beautiful. This is actually the replacement because the first one was worn out by people walking through the building. 

So were the Dutch intending to settle is a subject probably for another day but I will say that in introducing each of these lands what Blaeu then did, and I’m operating in a kind of world of publishing rather than in a world of real exploration here, what Blaeu was doing was creating a public for New Holland, Hollandia Nova and the intention to settle became something that would come up. In 17 ... as late as 1718 I believe there was a scheme by a scientist who was driven by the beautiful climate that he’d already ... that he had studied and understood so that was last time, a hundred years after the ... nearly a hundred years after the example that you were talking about. But really no undertaking of the kind that was launched in North America succeeded in Australia. There were a few lost sailors and there were a few voyages that lost entire ships but we’ve got really no evidence of a serious attempt and of course they had the difficult pro ... the memory of Batavia where they lost dozens to a crazy person so really it was not viable for them.

The difficulty was ... and if you were a Dutch cartographer coming to the south land and you were not meant to approach the west coast is that you really didn’t spend much time there. Even Tasman only spent a few days mapping southern Tasmania and then a few days mapping all of the east coast of ... that he did of New Zealand and it’s disappointing from that point of view because you know the longer you’re there the more chance you have to trade with someone so in essence the Dutch never really found someone to trade with and people say oh that’s because the Aboriginal culture didn’t trade. Of course that’s nonsense, Aboriginal cultures traded but this particular group were looking for short-term, low-volume, high-mass, high-value profit and this was not the VOC, the Dutch East India way. 

So here’s an example of wall maps that you will find in art and in this case the atlas there is the Clinker Atlas. There’s a reduction of a similar atlas here and the map from the Clinker Atlas is out there as well at full size so you can ... you saw from the previous image with Peter Barber that that’s a very big publication. This is an example of how it would be hung. 

So you saw the subject of our talk tonight, New Holland right in the centre of the map. In the Archipelagus Orientalis is front and centre and Blaeu is selling maps and it appeared that way, that this was an important publication. However because of the lack of colonisation and the lack of a trade and return on investment his map of New Holland never ... was hardly repeated. Even though it was produced a few years after Tasman it was hardly ever repeated in that view for the entire period of Dutch mapmaking all the way through to Cook. 

Here’s an example. Typically you’d find it at the east of an East Indian map or the north as it’s orientated in this way. You might find it on the you know the west coast might be on the right-hand side of a map of the Indian Ocean but you rarely found it in the centre of a picture. And it didn’t ... that plate never replaced any view like this in a Dutch atlas for the next 50 years so it really never became the important part of Dutch iconography other than as a way of saying this is where we’ve ... we have exceeded our previous explorers, the Portuguese, the Spanish or the Romans if you like. And the prize was always that cartouche as you saw on the map itself, was simply the prize in the centre of the frame which was the trade privileges, the licence to trade, the power to trade. 

So for me ... this is the map that emerges for me as an important map. This is the map as it was photographed three years ago and you know you can see there’s some damage there. The amazing thing about this is it’s the only example of a wall ... Dutch wall map to have survived to my knowledge with the rollers still attached. We’re still trying to determine whether they were the original rollers or not. The only way a wall map is going to survive is in a giant atlas and three were presented to royalty in the 1660s. 

There are a lot of details on the map which we won’t go into today but these are details that don’t appear on other maps so this map provided the first sighting of the Zeen on the west coast of Tasmania. It’s what’s not on the map that is really interesting from my point of view. The Dutch were interested in describing ... if you remember the previous image it had letter press text around the outside. This was describing all of the discoveries in the East Indie. There wasn’t a single reference to Australia except that it was discovered in 1644. All the rest is really detailed, this is the section on the Maluku Islands and it’s all of the description you could possibly want about the value of sandalwood or cinnamon or cloves and how to use them, how to harvest them and so forth. And we’re still in the process of translating each of these. There’s a Dutch, a Latin and a French.

And New Holland would have disappeared really but it became an inspiration. For me this is the map that everybody uses to remember Blaeu’s map with which is the map by Feveneau. Feveneau, French travel writer, former diplomat produced the map to illustrate the horrors of the Batavia wreck in 1629 so he used the map retrospectively I suppose or to illustrate an earlier story and called it New Holland but of course when the Batavia was the Batavia it w ... there wasn’t ... no New Holland but what was interesting is that this story was repeated and repeated and repeated through many editions into the 1700s and by now you have the Dutch ... the French and the English replacing the Dutch in the East Indies and vying for power in the Pacific and the Indian Ocean. And this is the Emanuel Bowen map which I think should be included in the cell exhibition because it includes great advertising for the Australian continent, saying that it’s the most productive and the most beneficial climate and so forth. 

And there’s another French version, in this case starting to extend further into the Pacific, people trying to get an idea of where that east coast might lie. But again as I say this is the view that we have and this in a sense is the archetypal archipelagus, it gives us the view of New Holland that the Dutch had, not a colony but an idea and Van Diemen’s idea really and unfortunately as a result of him expiring early perhaps there would have been a third voyage and something else might have happened to change the course of history but it wasn’t to be. 

We now have the map and the challenge for us is what to do next with it. It represents many difficulties both on the front of the map and behind the map and I’m going to hand over to Denyl to explain some of this. Thank you. 


DC:      Thanks Martin. In my role as Assistant Director Preservation at the Library I have been the project manager for the conservation of the Blaeu map. Preliminary conservation work was completed in 2013 in preparation for the display of the map in the Library’s exhibition, Mapping Our World. Many creases were flattened and the flaking and fragile paper was consolidated to enable the safe display and handling of the work. This is one of the overlay images that we created showing the areas of loss, creases and lifting fragments when the map first arrived at the Library. 

As we know the map was successfully shown in the exhibition and many of you may have seen it during its display however a question remained unanswered, corrosion of the green pigment of the map was thought to be affecting the supporting paper and that this posed a danger to the long-term preservation and display of the map. Verdigris is a brilliant blue-green pigment which was applied as a decorative element to embellish the surface of the map. Over the last 350 years the pigment had corroded to become a brown crust along the coastline of the surface as you can see in this photo, damaging the paper underneath and surrounding the pigment. Thanks to the generosity of NLA supporters and map lovers we could thoroughly investigate the verdigris and this next stage of the Blaeu map stabilisation could now go ahead. 

The Grimwade Centre for Cultural Materials Conservation at the University of Melbourne has access to state of the art laboratories that could be used to analyse and identify the material components of the map. The Grimwade Centre conservators would use this information to develop and test possible conservation treatments which would be considered by Martin and the preservation team before we start on the most appropriate treatment. This is the Grimwade Centre team visiting the map at the Library in March this year kick-starting the conservation project. 

So the Blaeu map began its journey south to the University of Melbourne. This was almost a project in itself, preparing the fragile map that was secured in a travelling frame inside a large crate. Nothing could be in contact with the fragile surface of the map and the crate was placed on many blankets inside the truck to minimise vibrations as much as possible. After it arrived in the University of Melbourne facilities the Grimwade Centre team had to wait 24 hours for the crate to acclimatise before they could open it. You can imagine our relief when we found out that the map had arrived unchanged. While we were preparing to move the map NLA Grimwade Centre to ... and NLA ... NLA and Grimwade Centre teams began extensive research into the map. Martin prepared research into the history, materials and techniques of Dutch mapmakers of the 17th century, my team prepared a package of information about the initial treatment that was undertaken on the map and the Grimwade Centre teams began researching possible treatment options and analytical methods. 

Now that the map was in Melbourne the first thing to do was document it. The Grimwade Centre team have developed a grid system to identify areas of the map so that they can identify specific locations for testing or during the treatment. Documentary photography was carried out using a range of photographic processes and flipbooks have been produced following the grid system. They will be annotated as the treatment progresses. The rest of the images and photo micrographs in my slides comes from the Grimwade Centre’s documentation. 

The photography includes visual light images such as the one on the left, images with magnification, breaking light images, ultraviolet light that show the varnish and extent of the original paper in the central image and infrared photography that show the original printing ink shown on the right. A comprehensive analysis of materials had been undertaken by Dr Caroline Kyi, Dr Stephanie Anne Alexander and Vanessa Kowalski identifying the composition and properties of the pigment and the varnish. Caroline and Stephanie have also made a significant exploration into the degradation mechanisms of the copper-based media on paper. The conservation scientists began their analysis of paper by investigating the pigments on the surface of ... before testing the varnish solubilities using fresh verdigris and the vial of scraped-off old varnish that had been kept from the initial treatment of the map. 

The results of the pigment analysis were fascinating. A pigment believed to be verdigris was applied along the coastlines. It is now fully degraded to black copper oxide, there is no verdigris left on the surface. It appears that lead white was used in association with the verdigris to highlight the small ship features throughout the ocean. The blue-green pigment used in the cartouche has been identified and is thought to be malachite and this pigment is more intact compared to the verdigris. Two distinct red pigments have been used, a lighter bright red which tested positive for mercury, suggesting vermilion on cinnabar and a darker red ochre pigment used in the cartouche. 

Gold leaf was used to highlight small details in the cartouche on the ships and on some lettering. The text of the banner title appears to have between fully gilded at some point. Gold is embedded in the paper fibres across all of the letters. They will use x-ray fluorescence to confirm that this is gold at their next opportunity. The varnish was a good match for mastic. Upon close examination it appears that its application was early but not contemporary with the assembling of the map. Incidentally they have found that 37 pieces of paper have been assembled to create the map. They have found that the original lining is more extensive and intact than first thought however this will be confirmed when the map is viewed from the back. 

Now that we understood the material components of the map the Grimwade Centre paper conservation team led by Libby Melzer began testing possible conservation treatment approaches in-situ but testing media solubility and varnish removal approaches. Initially these tests were undertaken separate to the map itself before direct testing of the map. It is proposed that the conservation of the Blaeu map will be carried out in three stages. Stage 1, varnish removal and the surface stabilisation allowing the map to be turned face down. Stage 2, removal of backings as determined appropriate, flattening and support stabilisation. Stage 3, infilling of losses, visual reintegration and preparation for mounting and storage. 

This gives you an indication of the complexity of the conservation treatment for the Blaeu map. For the varnish removal proposed in stage 1 there will be three different techniques used depending on the sensitivity of the media and the paper. The techniques use poultices of xanthan gum or clue cell G [? 30:52] or using a combination of ethanol and white spirits on different zones of the map. Extensive research has been carried out on these potential products, particularly into their ageing properties. While xanthan gum is a relatively new product for conservation purposes it has a long history of use in the food and pharmaceutical industry and is used as a stabiliser in ice-cream. 

The decision-making process for the map is also complex and approaches cannot be finalised until earlier treatments are completed, particularly for stage 2 work. For example the removal of the backings is still under consideration. Once we have been able to turn the map face down the backings will be assessed and fibre identification will be carried out by textile conservator, Marion Parker. Then we will be able to weigh up different options to work out the best way to proceed, that is, should the linings be removed and replaced or retained on the back of the map? 

In preparation for the treatment Grimwade Centre conservators have developed an extensive risk assessment outlining the benefits and potential risks with each aspect of the treatment. For example the cartouche is particularly fragile. The media could move if the cartouche is consolidated using a brush and as a result this work will be carried out under a microscope. At this stage the varnish cannot be safely removed from this small area and will be left on the surface. 

The treatment proposals are also based on the results of testing for copper mobility, also known as free copper. It is the free copper that causes instability of verdigris, changing the colour of the blue-green pigment to brown and causing the structural and aesthetic damage to the paper support where it can literally eat through the paper. The results of this testing were crucial as it would determine whether free copper is present on the surface of the map and if it is what treatment will be used to de-acidify the paper and remove the free copper. Testing has not found any free copper except in a few small areas where there is green ... still green pigment. These areas will not be treated with any moisture to prevent migration of the copper into other areas and the possibility of binding these remaining small quantities with modified gelatine are being explored. 

We are also discussing the possibility of accessing conservation treatment reports of other Blaeu maps. These treatments were completed some time ago so it would be interesting to see what’s happened to the maps since their conservation treatment. We have receive the condition report for a similar map, the wall map of the 17 provinces held in a castle in Stockholm. This map was treated in 1998 using a method developed by the Dutch National Archive that included both removing the lining and wet treatments. In hindsight the conservator who undertook the work and sent through the treatment report comment that it ... she didn’t know whether she would do it the same way now given the high risk of wet treatments but the map has continued to remain in stable condition. 

One very promising aspect of the testing is the response of the media on the surface of the map. All of the colours have saturated out and there has been a significant visual improvement which has got us very excited about what the map will look like after the treatment. Most importantly from the conservator’s point of view if we go back to the images of the testing the gold flake has remained on the surface right at the top of the mast. It shows that the treatment is successfully removing the varnish but not damaging surrounding surface. There are several months of treatment ahead of us however our thoughts are also turning to what happens to the map after the collection of the conservation treatment. As part of the project we will look at a storage system for the map which will allow us to safely store in the stack and move if needed. We are also keen to display the map in our Treasures Gallery and we’ll be sure to let you know when you’ll be able to see it again. Thank you. 


DC:      Martin and I would be happy to answer any questions if you could just wait for the microphone for the benefit of those wearing their hearing loop. If there’s any ... maybe not ... oh we have someone in the front row? 

A:         Hi. What was the fabric in the map originally? 

DC:      The linings? 

A:         What was it actually painted on, the map? 

DC:      So it’s painted on paper and then there’s two layers of fabric but we haven’t worked out what the fabric is yet because we haven’t been able to turn it over. 

A:         Okay. 

DC:      Yeah. 

A:         And was the varnish a way to preserve the map when it was installed on the wall or wherever it was? 

DC:      Yes. Similar to what you would do with a painting of the era, yeah, that’s why you would varnish it, yeah. It’s not particularly common for paper conservators to work with varnish so that’s why we’re working quite closely with paintings conservators because they deal with it all the time. 

A:         Thank you. 

DC:      My pleasure. And to ... I think that’s everyone. 

CVD:    Well thank you all for coming this evening and thank you to the National Library for having me here and to Martin for organising my visit. I’m going to be talking today about this map which was made in about 1491 by a cartographer named Henricus Martellus, German cartographer working in Florence and we have as we’ll see other works by him from the latter half of the 15th century. This map surfaced in Switzerland in the late 1950s, there was a flurry of interest in it, it was recognised as being both a genuine product of that era and of great importance. It was sold and anonymously donated to Yale University where unfortunately little had been done with it because most of the text on the map as we’ll see had faded to the point of being illegible so it was this great unstudied and unstudiable object and I’ll be talking about a way that we were able to enable studying of this map. 

But I’ll give some background by talking about the development of Martellus’ world maps from Ptolemy’s Geography to the map that we’ll be focusing on, the one at Yale University. Ptolemy’s Geography was written in the second century AD in Alexandria, Egypt which was then culturally Greek. It consists of an enormous database of about 8,000 place names with their latitudes and longitudes and that’s how Ptolemy organised space, by latitude and longitude. That seems obvious to us now but when his work was rediscovered in the late 14th and early 15th century that idea of organising space by latitude and longitude had been lost. 

So the work consists of this database of place names with their coordinates and then instructions for making maps. And I want to look at the structure of a Ptolemaic world map. So here’s a diagram that Ptolemy offers his ... for his so-called second projection for making a world map and we’ll just go through some of the details of that. We have here the equator, the southern limit of Ptolemy’s world map is the Tropic of Capricorn at about 23.5 degrees south. He indicates the Tropic of Cancer at 23.5 degrees north and the northern limit of his map is the Arctic Circle at about 66.5 degrees north. 

Ptolemy only shows 180 degrees of longitude, that is half of the earth’s circumference and he was very conscious of the fact that he was only showing half of the earth’s circumference. So if you fix the shape of this diagram in our minds I’m going to switch now to a map made on that projection and this is a world map in a manuscript of Ptolemy’s Geography that was made by Henricus Martellus, that is made by the same cartographer who made the map that we’ll be focusing on later. We don’t see a Ptolemaic map every day so we’ll orient ourselves a little bit. Here's Europe, you can see the ... Italy sticking down there, the Mediterranean so of course that’s Africa, the Arabian Peninsula and then moving eastward we have Asia and I want to emphasise that Ptolemy does not know the eastern coast of Asia so he doesn’t depict any water to the east of Asia. And then the Indian Ocean with the Island of Taprobana which is usually identified with Sri Lanka. 

Henricus Martellus, this cartographer I’m talking about, made other world maps that survive. He produced manuscripts of Ptolemy’s Geography with world maps like the one we just saw, he also produced island books that were illustrated with maps so text about the islands of the world, each illustrated with a map and some of the manuscripts that survive of the work include a world map like this one here which is in the British Library. If we compare a Ptolemaic world map ... Martellus’ Ptolemaic world map ... with that in his island book we can see ... and I’ve indicated there some points that are on the same meridians so that we can compare the east to west extent of the two maps ... we can see that the map in the island book shows us more of the earth’s surface, the Ptolemaic world map shows 180 degrees of longitude, the world map in the island book shows 240 degrees of longitude and what he’s added is more of eastern Asia. We now see water to the east of Asia, that is we see the eastern coast of Asia. 

And if we compare the world map in Martellus’ island book with that at Yale we can see that the world map at Yale shows even more of the earth’s surface, it shows 280 degrees of longitude and I think it’s quite remarkable that in the cartographic career of one man his world maps went from showing 180 degrees of longitude to 280 degrees of longitude. That’s a tremendous increase in the course of one person's career. And what he's added of course is what we might call some of the proto-Pacific, water further to the east of Asia going all the way to Japan. Where does all this new information come from? It comes from Marco Polo and in fact it’s quite easy ... when we ... when I was able to study the map it was quite easy to see where Ptolemy’s information leaves off and where the information from Marco Polo was added. 

So what reasons do we have to think that this map is important? What caused this flurry of interest when it surfaced in the late ‘50s? Well there’s three good reasons to think that the map is important. The first is that it very likely influenced Christopher Columbus’ thinking about the shape of Asia and the location of Japan. That’s a very bold statement, the idea that a specific map influenced a specific explorer. What I want to say is not that this specific physical map was in Columbus’ hands but that another ... this map or another map made by Martellus was seen by Columbus at some point. 

There’s a few pieces of evidence to support that bold claim but I’m just going to emphasise one which is the orientation of Japan and the principal axis of the island runs north and south and Columbus’ son in his biography of his father said that my father, Columbus, would certainly have discovered Japan had he not believed that the island’s principal axis ran north and south. There’s no other contemporary document, neither text nor map, that shows Japan with that orientation so that’s fairly strong evidence that this cartography influenced Columbus.

Another aspect of the map’s importance is that it influenced Martin Behaim, the German cartographer, in the creation of his terrestrial globe of 1492 which is the earliest surviving terrestrial globe. We have textural evidence for earlier globes but this is the earliest one that survives. And we can see that emphasis in southern Asia. I show these ... this huge peninsula, southern Asian peninsula on both the map and the globe ... on both the map and the globe the information in this area comes from Marco Polo. Marco Polo don’t give latitudes and longitudes, he says city X is 15 days’ journey from city Y and so creating a map based on Marco Polo involves a lot of extrapolation and it’s supremely unlikely that two people would extrapolate that data in the same way so this is good evidence for influence of Martellus on Behaim. Just to look at one other example, both the map and the globe show this enormous triangular peninsula jutting eastward from Asia and in both cases it does show just north of the tropic which I have indicated with the arrows there so the outlines of Asia are very similar on the map and globe. 

Third reason, third aspect if you will of importance of Martellus’ map at Yale is that there are striking similarities between that map and Martin Waldseemüller’s 1507 world map and it seems that Martellus’ map served as a model for Waldseemüller’s. And those similarities were recognised when the map surfaced but it was not possible to explore just how close the relationship was but I want to show a few of the details of that similarity and a few words first about the world map of Martin Waldseemüller. It’s a long wall map and as Martin said wall maps do not survive well. This one survived because the 12 sheets on which it was printed were put into a codex by a German humanist. 

It’s famous for including the new world and it’s the first map to apply the name America to the new world and it was for that reason that the Library of Congress in the US had been so interested in acquiring it for so many years and finally did so. The two maps have similar dimensions, they’re both large wall maps. Both cartographers take advantage of the large margins supplied ... left over as it were by the projection to include long text blocks in the lower left and right corners of the map. The shape of north Africa is very similar on the two maps. That’s no surprising because in both cases the shape of northern Africa comes from Ptolemy's Geography but the outline of eastern Asia is strikingly similar on the two maps, more so in fact than the similarity between Martellus’ map and Martin Behaim’s globe. And in particular both cartographers place the island of Japan at the eastern edge of the map. 

So there’s these similarities of layout, similarities of geographical outlines but how close is that relationship? Well the problem with the Martellus map as I mentioned before is that almost all the text on it has faded to eligibility. Just to look at a few details on the map. This is north-eastern Asia in natural light so as the human eye sees the map and hardly any text is visible, you can perhaps see there are a few banners that indicate the names of regions but given the map’s size we would certainly expect it to include descriptive text. And there are in fact some descriptive texts that are just visible if not exactly legible in north-eastern Asia and this one I’ve indicated here is the most legible in north-eastern Asian which is to say not very legible at all. The island of Japan in natural light, we don’t see any text on it. We would expect there to be descriptive text not only again because of the size of the map but also because Martellus made another map of Japan in his island book that has descriptive text on it so he had descriptive text at the ready as it were for Japan. 

So how can we get at the text on this map? The solution is what’s been called multispectral or hyperspectral imaging. There are distinctions between those two terms but the people who do this imaging are not consistent in the way they apply them and I think there’s a good argument to be made for just calling it spectral imaging. But the basic idea of spectral imaging is that one takes a series of digital images of the object at specific frequencies of light and one might use say a dozen different frequencies of light ranging from the ultraviolet through the visible spectrum into the infrared. And the idea is that one processes the images, combines them in a way ... such that the information revealed by each of the images is combined into one image so that one can see as much as possible of the object. 

This technology’s useful for recovering texts from documents damaged by fading as in the case of the Martellus map, water, fire, overpainting, palimpsesting, when a text is written in a manuscript, scraped off so that the parchment can be reused and then new text is written over the text that had been scraped off. The famous example of that is the Archimedes palimpsest which was a ninth century that had unique texts from the mathematician, Archimedes, that were scraped off and it was remade into a prayer book. Using multispectral imaging it was possible to recover those texts that had been scraped off and finally the technology’s also useful in recovering texts on manuscripts damaged by wrinkling. 

And so we had some good indications that the Yale Martellus map was a good candidate for this technology. Going back to north-eastern Asia, again the image looks pretty muddy and that’s really how the map looks in person, it’s difficult to see anyone ... anything but when the map surfaced some ultraviolet images were made of the map and I’m going to switch now to an image that shows exactly the same area on the map. This was one of the most exciting images I’d ever seen of a map. What it shows is that there’s text everywhere in north-eastern Asia and this really made me very, very interested in doing everything in my power to get at that text and see what it had to tell us. 

Comparing south-eastern Africa in natural light and again ultraviolet light, the ultraviolet light reveals that the river system of the Nile very curiously extends very, very far to the south in Africa which is something quite surprising. And again because multispectral imaging makes use of ultraviolet light the fact that ultraviolet light would reveal this much about the map suggested that even more could be revealed by multispectral imaging. Looking at the south-eastern tip of Africa in natural light I’m now going to switch to showing that same region in infrared and we can see that the river systems and the little indications of trees near the rivers and also the names of the rivers become more visible. 

One last example, in this natural light image we really can’t ... just barely see that there might be some text in the area I’ve circled and in ultraviolet light the text appears reasonably well so these are all good indications that multispectral imaging would be good technology to try in this map. I ... together with my colleagues in the Lazarus Project I managed to secure a grant from the US National Endowment for the Humanities to make multispectral images of the map and we went to Yale in August of 2014. 

Here’s the team, that’s me on the left, that’s ... next to me is Ken Boydston, the CEO of a company called Megavision which makes very good equipment for doing multispectral imaging, in the middle it’s Roger Easton of the Rochester Institute of Technology who did most of the processing of the images that we made of the map, that’s Michael Phelps who directs a project called the Early Manuscripts Electronic Library which is doing multispectral images of the palimpsest manuscripts at St Catherine’s Monastery in the Sinai which has the largest collection of palimpsestic manuscripts in the world, and on the right that’s Gregory Heyworth, the Director of the Lazarus Project who had the idea of developing a portable system to do multispectral imaging. Many institutions only have half a dozen good candidates for multispectral imaging so it would make no sense for the institution to acquire the equipment to train someone to use it so Gregory realised that what’s needed in many cases is a portable system that goes to the institution. Not only does the institution not have to purchase the equipment but the precious objects don't have to travel.

Here is the Martellus map out of its case that ... removing it from its case was an undertaking performed by the curators at the Beinecke Library, a very delicate operation. You can see that the map is sitting on an easel which we brought with us. The easel is tilted back at about 10 degrees so that the map is held in place by gravity. When doing multispectral imaging it’s very important that the spatial relationship between the camera and the lights remain constant and so the easel is able to move the map left and right and up and down so we organise things so that the camera and lights would remain stationery while the map moved in front of the camera and lights. 

Just another view of the map on the easel. The two bars on the camera, the metal bars are the laser-focusing system. And just to look a little bit more at the equipment. On the top on the tripod, the high tripod there is a multispectral LED light source. The way multispectral imaging used to be done is that one would shine a very bright white light on the object. White lights contain all the frequencies and then one would put a filter in front of the camera. Well there are two disadvantages to that. First of all the object is subjected to too much energy and second, any filter involves at least some distortion. So the more recent technology, what one does is one shines only the frequency on the object that one is interested in so the precise frequency, it involves an absolute minimum of energy. 

In between the lights and the map in this case is a diffuser which ensures that the light arrives evenly distributed on the object. For purposes of imaging we divided the map into a grid of 55 overlapping tiles and so these were the parts of the map we were imaging as we moved it in front of the camera using the easel and of course they were overlapping so that later we could digitally stitch the tiles together. When doing multispectral imaging one spends a lot of time in the dark with these different-coloured lights flashing and one really hopes that one’s colleagues have some good stories to tell. 

So that was a lot of preamble to get to the results. So let’s begin by looking at this text block in the lower right-hand corner of the map. We’ll zoom in on it here. Here it is in natural light so this is how it appears to the naked eye. One can see that there are letters but they’re ... trying to put them together into words is very challenging indeed. Here it is in infrared light. One might argue that a few more letters are perhaps visible but it’s really not much help at all. For some reason the ... just the ultraviolet light by itself was of no help whatsoever and there’s the multispectral image. So the entire text is suddenly in ... quite easily legible. And one thing I light about this sequence is that both infrared and ultraviolet light have been used in the past to try and recover text from damaged manuscripts and this shows very clearly how much more powerful a tool multispectral imaging is. 

Looking at some other details from the map, zooming in on the Alps. Again the surface of the map in natural light looks just ... it’s a murky mess and that’s ... what you’re seeing there is how it looks in fact and then the multispectral image, there’s place names everywhere. I found this image to be quite surprising. I suspected that there would be many place names in Europe but I didn’t suspect that they would be quite this dense. Moving to northern Asia, again the natural light image, multispectral image of exactly the same area and here we begin to see at the left and right some of the longer descriptive texts that I’ll be talking about later. 

We’re turning to Japan. Again in natural light it’s ... couldn’t really see any text there at all. Here’s the multispectral image, text comes out quite nicely and it looks fantastic until one starts trying to read it. And many cases in this map, the multispectral imaging makes the text visible but it’s still considerable effort to read it but in this case as I started to try and read the text I noticed that parts of the words were missing. And I wrote to Roger Easton who was processing the images and asked if there was another processing technique he could try to try and bring out that missing text and sure enough he was able to do that. We can see now that the letters go all the way to the edge of the island. 

And this brings us to the importance of image processing or Roger, can you give it one more try? While I was working on the map I had the luxury, and it was a luxury, of being in constant contact with Roger and he would send me processed images and I would study them and I would send him back Powerpoints highlighting the areas that I was interested in and asking him if these was anything else he could try and we ... in order to do this project we spent 10 days at Yale making the images but making the images is let’s say 15, 20% of the effort, the processing is where the magic happens. And I want to go through some cases that show how important the processing was on the images of the Martellus map. 

Going to India we can see even in the natural light image that there’s some text near the top of the image. It becomes much clearer in the multispectral image but the area I want to focus on is here. We don’t see any text there in this processed multispectral image nor in this one processed using different techniques, nor in this one processed again using different techniques and yet with a different technique this text suddenly appears. And this happened to us quite a few times with this map and what we learned ... well what we suspect is that the texts were written using different pigments and the pigments respond differently to light. And what we came to conclude is that whenever there was an area on the map that seemed to be blank there was probably text there, it was just a matter of finding the right technique to reveal it. 

Going ... returning to north-eastern Asia, again the natural light image, multispectral image and I want to point out here that ... we’ll talk about the text in just a second but the multispectral images don’t reveal the text, they also reveal the mountain ranges and the images of cities so I’ll go back and forth. So here’s the natural light image, the multispectral image, you can see the mountain ranges very clearly now and the images of cities in the north. The image also reveals these two texts here quite well. In these other three cases we can see the lines to guide the writing of text but the text itself is not really visible and this is another case where the text was written in a different pigment and a different processing technique was necessary in order to reveal it. And this physical fact about the map, the fact that the texts were written in different pigments is something that one would never begin to suspect looking at the map in natural light. 

So we were able to read most of the text on the map. What can we do with that? Well one thing we can do is see just how close the relationship is between the Martellus map and Martin Waldseemüller’s world map of 1507. So I’ll go through several examples of texts in the same locations on the two maps so beginning in north central Africa here's a natural light image of the detail, the multispectral image makes it legible. There is a corresponding text on Martin Waldseemüller’s world map, a text in basically the same location. Here are the transcriptions. I’m not expecting you to read Latin. What I’ve done is underline the words that are the same in the two the texts so it’s the words that are not underlined that are different and what we see is that the texts are very, very similar, it’s essentially the same information in the same location. And just to translate, the text says here there are great wildernesses in which there are lions and big leopards and many other animals different from ours, that is, different from European animals.

Moving a bit to the southeast in Africa here the natural light image, multispectral, now it’s legible. There’s a text in the same location on Waldseemüller’s map. Transcribing, you can see there are two words in the upper text that are not underlined so just two words of difference between these two texts. And again just to translate, he’s talking about the serpent that’s born here that causes the ground to smoke and I’m not quite sure what that means but that’s what the text says. Moving out into the Indian Ocean there’s a cartouche here that’s quite illegible in natural light, the multispectral makes it legible. There’s a cartouche in the same location on Waldseemüller’s map, now we can compare them. And we see here there’s just one word of difference between these two, there’s one word in the lower text that’s not underlined. So the same information in the same location and he’s saying here is seen the Orca sea monster that’s like the sun when it shines whose form can hardly be described except that its skin is soft and its body is huge. 

Moving into Asia, again natural light, multispectral. One interesting thing about this case is that not only are the texts in the same general area of the two maps but in both cases they’re between these two mountain ranges. And again one word of difference between the two texts so the same information in the same location and he says here there are monsters similar to men but with ears so large they cover their whole bodies with them. So these are the famous Panotii that go back to classical antiquity, one of the monstrous peoples that people believed in. Going to south Asia, again natural light, multispectral, the text in the same point on Waldseemüller’s map. Just one word of difference, the first word. He says here there are many animals different from our animals, again different from European animals. 

So the world map by Martin Waldseemüller is the most expensive map ever purchased, $10m so did the Library of Congress pay that much money for a map that’s simply a copy of another? And fortunately the answer is no. So I’m now going to look at some cases in which Waldseemüller modified or departed from what he found on Martellus’ map. I mentioned before that the shape of North Africa is very similar on the two maps. Well the shape of southern Africa is entirely different. On Martellus’ map I show the details on the right there. On Martellus’ map South Africa looks like a shoe more or less whereas on Martin Waldseemüller’s world map its shape is much more familiar and correct so this is one just quick indication that Waldseemüller did not simply copy from the earlier map. 

I want to look at a chunk of coast on western Africa and I’ve chosen an area that’s north of that very strong difference in the outlines of the continents so this is an area where we might expect the coastal place names to be the same. Here is the detail of that area on Martellus’ map, the place names are not just illegible but invisible. The multispectral image brings them out quite nicely and then comparing that same chunk of coastline with Waldseemller’s map we can see immediately that there are more place names on the later map and if we compare them closely it compares that there’s not a single one that’s the same. So it turns out that Waldseemüller was using an entirely different source for the place names on the coast of Africa. 

Moving to the Arabian Peninsula, this is one of the few areas on Martellus’ map where some of the place names are actually legible in natural night. Nonetheless the multispectral image helps a lot. If we compare the same area on Waldseemüller’s map, in both cases all of the place names come from Ptolemy's Geography and we might expect that since Waldseemüller had Martellus’ map in his workshop he would simply copy all those place names. He did nothing of the sort. I’ve here circled the place names in the middle part of the Arabian Peninsula that appear on Martellus’ map that do not appear on Waldseemüller’s and here are the ones on Waldseemüller’s map that do not appear on Martellus’. So Waldseemüller for whatever reason made an entirely independent selection of place names for the Arabian Peninsula using the same ultimate source that Martellus had. 

Moving to the northern ocean we have here an island with cartouches above and below and on Waldseemller’s map we don’t have that island, we just have a quite large cartouche and in fact it is a case again where Waldseemüller did not copy Martellus. We have this island and the descriptive text on Martellus comes from Pliny the Elder whereas on Waldseemüller’s map there’s no island and a text from Pomponius Mela. So again he did not simply copy everything he found on Martellus’ map. Moving into the Indian Ocean we have here this text ... cartouche which is now legible and the text is about a fish called the Narco that makes a fisherman’s arm and whole body numb even through the fishing pole and line and there is no corresponding text on Waldseemüller’s map so again he did not simply copy everything from the earlier map.

Looking at the texts in the lower left-hand corners Waldseemüller’s text is substantially longer, partly because he’s talking about the discovery of the new world which had not occurred when Martellus made his map but I find it very interesting that even though the two cartographers are talking about basically different subjects Waldseemüller nonetheless copies that whole final phrase from Martellus so even when he was thinking about a different subject he still had the earlier map in mind, he still consulted it to see if it had text on it that might be useful to him.

So a few conclusions. One of the very exciting things about this project is that this map which and been unstudiable is now studiable in all its aspects and all of the images of the map will be made freely available on the internet and that includes not only the processed images but also the raw data. And what’s good about that is that if new techniques are developed for processing the data in the future that we’ll pull ... we’ll be able to reveal more information about the map, the raw data’s already there, the map will not have to be disturbed, it won’t have to be pulled out of its case again.

And the other thing that I find very interesting is that the study of the map, the images of the map enable this comparison between these two world maps and we get to see how Martin Waldseemüller worked. We get insight into his workshop practices, what sources he used, what parts of the earlier map he used, which he didn’t use and that’s ... insight into workshop practice that for an early 16th century cartographer that we can’t get from other means. The cartographers from that period didn’t keep a journal explaining how they made their maps so in this case it’s by studying the maps themselves that we get insight into the cartographer’s techniques. Thank you very much. 


[End of Recording]


Download transcript 118.84 KB

Recent audio All recent audio