Ring o’Roses: minimap as itinerary Pt 1

[correction made – 18/05/2017. ]

The ‘mini-map’ contains just one structure (right) representing the Latin west, the only reference to western Europe anywhere in the map though so vaguely defined (as: ‘between the inner and outer seas’) that it seems to betray an indifference as great as we find in the rest of the manuscript.  If this roof is meant as a roof, Avignon or Peñiscola are reasonable possibilities; if as a flame, it could be any western beacon existing during the early twelfth to fifteenth centuries, or which was remembered to have once existed.  Porto Pí in Mallorca or La Lanterna for Genoa are two possibilities.

The Voynich map

Overall, the Voynich map (left) is fairly called a map, but what

The ‘minimap’ occupying the map’s Northern roundel – (not to be confused with the emblem for North)

now fills its northern roundel – the “minimap” (right) – is concerned with just one route through the eastern Mediterranean: from the Black Sea via Constantinople to Egypt.


It is not so much a ‘map’ as an itinerary: that is,  a nicer-looking version of that type of ‘back-of-an-envelope’ sketch you’re given by someone familiar with a particular road or journey.

The difference between ‘map as map’ and map as graphic itinerary is explained later; the chief point now is that they were not then, and are not now, ‘pure’ forms.

Keeping in mind that the map as now bound has east to the viewer’s upper left, it is evident that the three centres emphasised belong to the north, south and eastern side of the starry sea; their regions coinciding with those of the alliance which Ciocîltan calls the ‘Serai-Constantinople-Cairo axis’.

To the north, the Mongol Khans who controlled the Black Sea had capitals at Tana and at Serai. To the south, the Mamluk capital was Cairo. Between them lay Constantinople, serving as halfway house and diplomatic intermediary, its interest being not only revenue from the eastern trade but safety from the military ambitions of each principal.

Genoa and Venice were not important to the principals in political terms, but their citizens were essential cogs in the commercial machine: these Latins had the means and know-how to maintain commercial traffic, generate vital state revenues and provide military or naval assistance as wanted.

It was an alliance which certainly had a profound  effect on Europe’s history from the latter part of the thirteenth century to the end of the fourteenth, and provides the wider stage in which the Latin cities fought for preferential status from Serai and/or from Cairo. It also created the political and economic environment which permitted a Genoese ship to carry an eastern disease – the Black Death –  from the Crimea to infect Marseilles, Paris, Rome, London and central Europe. For these and other reasons, I have treated this section in terms of that alliance and its time.

Emblem for North – description

It has been clear for some  time that the emblem used to denote the map’s North is a geographical one: a site-plan, and that this place is beyond the Black Sea.  By 2012 I was considering Mtsketa; somewhat later, Tabriz.

The site appears as follows: roughly circular though flatted along the perimeter from East to South; partly protected by two arcs of palisade – one from approx due north to East; the other from NNW approx. though SSW approx.

Divided by two lines, coloured blue, the longer dissecting the space approx. NNW to SSE, and  the shorter  connecting the centre to a point on the perimeter SW. (I am speaking of the lines’ direction not  direction of flow).

Two ways connect it to the wider world, a smaller path southwards drawn reminiscent of a rutted cart-track, and what is fairly termed a highway or king’s road leading east. This is made broad, smooth, with even sides and edges –  and supported by an embankment.

Despite those palisades and roads, the drawing is constantly misread by newcomers as a ‘T-O’ diagram  –  so commonly that we must once again digress.

A T-O diagram shows a circular world entirely surrounded by the wide band of an external Ocean.  As a general rule, a T-O diagram is oriented towards the east (that is, drawn with east up) so its longer line of division runs directly north-south, separating Asia from the two smaller regions below:  Europe on the northern side and Africa on the southern side. This idea of the world – all of the world – required an absolute boundary, which was envisaged as that encircling Ocean.

No highways or cart-tracks ever were, or ever could be drawn leading out from the  T-O diagram. There was no-where for them to go: it represented the whole earth.

So while one might offer a metaphysical or a metaphorical explanation for the  roads and palisades, by default the emblem must be read  as a real place having just one minor point in common with a T-O diagram: the space is divided into three parts, one occupying half the space, and the other two equal to each other in size. We can’t even say with certainty that the lines now coloured blue were intended as water or whether, for example,  the Voynich manuscript’s painter just supposed they should be.

 (emblem for North) … Serai..Tana

That I have since concluded that the North emblem is probably a token for Tana or Serai is not due to theory, but now having more and better data  from the primary source’s internal evidence and from more recent historical and archaeological sources than were available five years ago.

I incline more to Tana than to Serai,  in part because medieval Tana (modern Azov) was near ancient Tanais (see composite illustration above), and partly because Tana is the more likely to have been the northern-most point of interest for those who made our present manuscript in the fifteenth century.

During the 1430s for instance, Pegolotti assumes Tana well known, adding no explanatory details when he writes …

“The road you travel from Tana to Cathay is perfectly safe, whether by day or by night, according to what the merchants say who have used it.”

and he offers detailed advice for the (apparently numerous) Latins setting out overland for China:

(detail) Cambridge, Trinity College MS O.2.48, folio 255r.    During the Middle Ages the word dragoman entered European languages: in Middle English as dragman, in Old French as drugeman, in Middle Latin as dragumannus, and in Middle Greek δραγομάνος.” – wiki

“In the first place, you must let your beard grow long and not shave. And at Tana you should furnish yourself with a dragoman. And you must not try to save money in the matter of dragomen by taking a bad one instead of a good one. For the additional wages of the good one will not cost you so much as you will save by having him. And besides the dragoman it will be well to take at least two good menservants, who are acquainted with the Cumanian tongue...”


I’d like to write more about the archaeological evidence in favour of each site, but for those researching the Voynich manuscript, I think the effort of absorbing that opinion will be quite as much as they need at present. 🙂

One might argue over whether the arcs of palisade are intended as natural or as man-made features but the  high road is certainly man-made.

It is the other path which takes one southwards to the minimap and its itinerary.

The North emblem and lead-in to the minimap.

Itinerary and/or map?

The difference in definition depends chiefly what the recipient expects. In theory, a map should use proportional distance between points, but in an itinerary, and though distances may be known, they need not be represented accurately, even in theory. Whereas the greater part of the Voynich map is meant for a map, this late addition is simply a graphic itinerary. The illustration (below, right) still has northwards to the upper right and eastwards to the viewer’s left).

Such an itinerary sets out, in order, the places relevant to your needs, omitting others which would be included in a map, or if you were a different traveller. As well as providing  basics, the itinerary’s drawings may be adorned with a few memory-jogging ornaments (literal or not). This is because having to memorise in advance a heap of place-specific instructions that won’t be used for some time  isn’t always efficient. Anyone who doodles will probably understand how mnemonics work.

I’m not sure whether I’m more embarrassed about how long it took before I realised that this section isn’t ‘map-like’, or more pleased to have finally recognised this distinction between the broader and older frame of the map and this late-added detail.  It explains why the ‘castle’ isn’t where I thought it should be.

In terms of the a ‘map as map’, the fortified  ‘castle’ (so called)  should have been at some place on the eastern Mediterranean coast: Aleppo, Tyre, Antioch, Ayas/Laiazzo or some such.  Realising, finally, that it was a token for  Constantinople came as a surprise.  But it was another of those errors resulting from unexamined premises. I had assumed the same principles informing the rest of the map also applied to this, despite its being a late addition, drawn in different style.

The reason that ‘castle’  (or more exactly one of its steep-descending flanks) occupies the itinerary’s mid-point is simple: Constantinople was the ‘half-way point’ in practical as in conceptual terms, and in terms of  contemporary politics, diplomacy, commerce and logistics. It may not be the geographic mid-point, but calculated as days at sea, and with the notoriously fickle conditions in the Black Sea, the difference in time might not have been great.

In any case here is the minimap, now adjusted so that North is up and east to the viewer’s right. (N.B. this required ‘flipping’ the map, and so reversing the script.)

I’ll use the rest of this post to consider the ‘castle’ in more depth, then turn briefly to the southern terminus (or beginning) of the itinerary.

In next post I’ll return to that featureless section  between the itinerary’s north and the chimneys of Cappadocia.

Galata and Constantinople

There’s really no way to explain briefly how this image works, so I’ll be content to demonstrate that even a modern ‘map-as-map’ commonly contains a mixture of literal and non-literal elements.  The latter are still more common in works of the medieval period.

Map and itinerary styles combined.

In the following tourist map of modern Istanbul, the literal elements need no comment, but were you to hunt the actual city for a long road whose surface was painted a uniform bright red, I doubt you’d find one. Nor a rowing boat half the size of a house.  Classing one detail as literal and another as not literal and/or not intended to be read to scale is fairly easy for us when we look at sources such as this.  Yet the habit  of most Voynich researchers has been to assume greater  (rather than less) literal intention on the part of the Voynich map’s maker and there has been a positive resistance to information equivalent to: “the blue road may not be intended to be read as literally blue” or “the rowing boat may be no more than an ornament for the picture” or “it is unlikely that the numbered ‘signposts’ were actually as large as one of Hagia Sophia’s smaller domes”.

As it happens, the most literal elements in both the Voynich map as a whole, and in its ‘minimap’ are topographic. The next most reliable in those terms are minor architectural  details of the sort which mark a structure as indigenous to one region or another.

As example, the map contains, in its section of road between North-to-NorthEast, a perfectly accurate juxtaposition of  massive crescentic dunes with the dominant architectural styles which one encountered in travelling east towards the Himalayas. (This section was covered in the map’s original analysis, published between 2011-12 and I use the same comparative images used then).

Photo of the Taklamakan/Taklimakan dunes courtesy of Brian McDairmant,  included in ‘Eastern Quadrant: the Taklamakan – and implications’, voynichimagery.blogspot (Monday, October 10, 2011); ruins of Khorasan by Steinmetz; curved roofs of Bhimilkali see ‘folv 86v Ways to the East – the desert road’ voynichimagery.wordpress.com (August 25th., 2012).

“The largest crescentic dunes on Earth, with mean crest-to-crest widths of more than 3 kilometers, are in China’s Taklimakan Desert.” US.Gov. Publications Service Centre.

Items that I read as being intended literally in the ‘castle’ image include that great tower at the rear; the square tower-like structures attached to the rear wall; the fortifications’ presence on all sides, and the fortified walls’ having openings very close to the salt-waters’ edge.

The steep-descending flanks, however, I believe to be like the tower in alluding to the “galata” though what we now call the Galata tower was then formally Turris Christi and the area in which it stood, Pera.  Below is a late fifteenth century Turkish image of Galata and its tower. For readers’ interest I compare it to the modern tourist map.

An earlier fifteenth-century image (1422) by the Florentine cartographer, Christoforo Buondelmonte, clarifies an interesting point: that where we’d be inclined to think of Constantinople as separate from Galata/Pera, contemporaries might see the two as parts of a single whole, separated merely by a  wider-than-usual inlet or harbour.

It is interesting to note the different form which the Turkish and the Latin painter give the great tower.

It is also possible that in the Voynich vignette the tower we see is not meant for the Galata but some earlier and other tower, though I think it more likely that the Voynich ‘castle’ is the result of adjusting some customary token for Constantinople to take account of current conditions and specifically of the movement of the commercial centre, and many of the Genoese, from the northern side of Constantinople to the market-site, Galata.

The two maps shown below illustrate this matter and the following quotation from Mango’s valuable paper may also prove helpful.

The maps show the foreign quarters, city by city, outside the walls.  They also demonstrate the fact that every street leading to the sea ended in a ‘sea-gate’ through the fortified walls. We’ve already noted (in this post) that the ‘castle’ appears to have faced the Horn rather than the Sea of  Marmara.

from the superb interactive map offered online by Nicolescu and Saffran.

The date of the Galata’s completion –  1348 – allows us to narrow the dates posited for the Voynich map’s last recension to  c.1348 – 1438, the first  being the date of the Galata tower’s completion and the other the lower end of the Voynich manuscript’s date-range according to the radiocarbon dating carried out at the University of Arizona.

Now here’s where I expect things may get a bit awkward, given that there are more computer science chaps in Voynich studies than persons with a background in  medieval history, palaeography, medieval literature or writings; art-history or iconographic analysis.  But anyway, here goes ..

Those steeply descending flanking walls or cliffs to either side of the ‘castle’ cannot be meant as literally as they appear. As I pointed out some time ago, it would be more than merely foolish to build a castle blocking the mouth of such a valley, and this is so in terms of hydraulics as in military terms.

In appearance they recall the bastions of Antioch and, again, the frozen cascades of Hierapolis, comparisons which might have been made intentionally, as ‘memory-joggers’.

The form and relationship of parts for Constantinople-with-Galata do compare with those of Antioch (below left) and both cities’ steep descents carried fortifications – those of Antioch also serving to carry water.

ramparts of Antioch – medieval








‘Milky’ Hierapolis.

Etymologies offered for ‘galata‘ have included the Italian calata (steep descent) and the Greek ‘milky’  γάλακτος (cf. amelgein). 

The first hardly needs justification; but the second could explain why an allusion to Hierapolis’ ‘milky’ but solid cascading tiers would serve as memory-aid. I have encountered a suggestion that medieval Greek used ‘milky’ to describe a shyster or slippery character… and while such questions are for the specialists, it isn’t difficult to imagine some proverb being coined about getting milked in Galata. It was the Genoese commercial centre.

Further, the fact that the exact centre of the itinerary is not filled by the ‘castle’ but but by the northern side’s ‘steep descent’  might imply that the person for whom the itinerary was made wished to land – not at Constantinople proper – but at Galata; the same may be implied by the prominence given the Latins ‘swallowtails’.  But altogether this detail offers an interesting suggestion that  Pera was known as calata and/or γάλακτος long before  official adoption of that name.

To end, an excerpt from Mango’s valuable paper:

“We begin with the Strategion region. Constantinople’s north shore continued to thrive thanks to alterations in trading patterns within the city that developed after the days when the Book of the Prefect was written. The growing role of Latin merchants in Constantinople accelerated a demographic shift in the city’s population, which had begun before the Fourth Crusade … and increased in the Palaiologan period. This resulted in a greater concentration of commerce on the Golden Horn, where the Pisans, Genoese, and Venetians chose to settle. [but] Commerce also moved over to Galata. In this late period, the Prosphorion was apparently no longer an enclosed harbour. Rather, Crusaders and others make repeated references to the Golden Horn itself as a harbour, served by skalai. There, in the late fourteenth century, the anonymous Russian pilgrim refers to “the large Basilike market near the wharves and ferry crossing to Galata.”(p.205)

and still with regard to the thirteenth to fifteenth centuries:

Myrepsoi, who dealt in spices and unguents, were located in the Portico of Achilles between the Milion and Chalke, silversmiths were still on the Mese, and textiles were sold in the forum of Constantine.”(p.207).

Maria Mundell Mango, ‘The Commercial Map of Constantinople’ from Dumbarton Oaks Papers, No. 54. pp.189-207.

Note: I am not in which sense ‘scalai’ is meant here, but in Greek nautical usage today, we have

Skala   Σκάλα
The port that is associated with the (highest) Chora. From the Italian word “scala” (stairs), as can be found on Santorini, Patmos, Astypalea and Sikinos. Skala is also used to indicate a quay on nautical charts.

.. this is long enough. Egypt can wait.


  1. By coincidence I came across a youtube video that reminded me of this post. They discuss navigation charts made by Marshall Islanders.

    Some comments really stand out in the light of the VM foldout. For example, the commentator says that the maps were not meant to be taken on the ship, but rather they were constructed to help one form a mental image of the area.
    They depict currents as “roads”, and those are given much more prominence than the islands, which are symbolized by mere shells.
    It’s not related to the VM, but I found it a nice and very accessible illustration of the kind of thing you talk about here.


    • Well there’s no direct link, only it’s a similar example of how a map doesn’t need to be a precise representation of the world. It’s tailored to the needs of those who made it. There’s a focus on itinerary and connections and roads. Also, the fact that apparently these things are designed to facilitate memory retention.

      Finally, in the video it is mentioned that these maps would mostly be used to teach others. Kind of as a visual aid to oral tradition. I think of the VM as something similar, an experienced traveller or navigator doesn’t need these things on paper 🙂


      • I see what you mean, but then again it’s pretty clear that the compilation was made to be taken along with the person. The map is the most worn and greasiest folio in the manuscript and is stained by water – maybe even saltwater, who knows?
        The botanical imagery and the ‘leaves and roots’ section is intended to assist identification and use ‘in the field’ (or more exactly, I’d say, in the market-site).
        The thing is made pocket-size.
        The presentation is not compatible with production from a Latin European scriptorium or atelier. It’s one of the viliores type.
        Most tellingly – because Latins learned to rule out the page before anything else – there’s no sign of ruling out AT ALL. That’s a huge deal: a real ‘tell’.

        I doubt it ever saw the inside of a nicely panelled library before the seventeenth century – but the sources from which it was drawn are another matter.


        edit – Postscript. Keeping in mind what Panofsky said of the manuscript, it may be explained partly by the overall look of it. I went hunting an explanation for what it was about the ms that led to his opinion that it was Jewish from ‘somewhere southern’ and with something of Kabbalah in it. We assume of course he is speaking of the pictures, but not necessarily so. I read with attention and respect the studies published by Beit Arie, and also a fine comment in a study of, specifically, early manuscripts of Kabbalah. One sees where Panofsky was probably coming from.

        One post is which may be of interest – especially with your close study of the Trinity College ms :-


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s