There’s a tiny error on this map. Made around the time of the New York World’s Fair in 1939, it shows the airports and air routes of the Soviet Union at a time when the USSR was keen to show visitors a country stitched together by aeroplanes. Follow the lines north until they run out and you arrive at a dot near the mouth of the Yenisei, labelled Dickson. But in 1939, it should ~ of course ~ be marked ‘Dikson’.

Adolf Erik Nordenskiöld named the harbour in 1875 after Oscar Dickson, a merchant from Gothenburg who’d paid for his foray north. Russian usage dropped the ‘c’ almost at once, and Dikson has been the official name since 1884 ~ which means on the day it was printed, this map was fifty-five years behind the curve.

You’ll have spotted this is an air map: Dikson is still the northernmost airstrip on any continental mainland, and from 8 December to 5 January it gets no daylight at all, not even twilight ~ but that’s almost a moot point.

I’m about to describe a fight between a cruiser, an icebreaker and two guns parked on a pier. The only aeroplane in what follows is a German floatplane, and even that gets wrecked before the shooting starts ~ but I liked the map, and I’ve been posting notes and tweets and whatever-it-is-bluesky does about Soviet mapping today, so this is the map you’re getting. It’s that little red dot, that deserves our attention.

In August 1942, Dikson was the scene of what is usually described as the only battle fought against German forces beyond the Urals. The Germans arrived by sea, certain they could take the land base without too much trouble, and they left, having been put firmly in their place – and without putting a man ashore.

BREAKING THE ICE

The German heavy cruiser Admiral Scheer left Narvik on 16 August 1942 for the Kara Sea. As one part of Unternehmen WUNDERLAND, her task was to find and sink the convoys using the Northern Sea Route along the Siberian coast.

Operation WONDERLAND was considered a first, key stage in establishing control over the western Arctic. As is often the way, Mother Nature didn’t attend the same briefings. Pack ice turned the Admiral Scheer back from the Vilkitsky Strait, fog hid the convoys her floatplane had found, and then the floatplane itself was lost in what might be called a ‘landing accident’. Her captain, Wilhelm Meendsen-Bohlken, also discovered his charts could not be trusted. Rocks stood where open water was marked, and shoals where the chart promised depth.

Maps, huh.

But then on 25 August, her lookouts sighted a single ship, the Alexander Sibiryakov, a Glasgow-built steamer of 1909 that had started life hunting seals off Newfoundland.

In 1932 the Alexander Sibiryakov had made the first single-season passage of the Northern Sea Route, finishing it under improvised sails after losing her propeller. The Soviet Navy had requisitioned her in September 1941 and given her a handful of light guns – which is something of a coincidence.

Alexander Sibiryakov, the Siberian gold merchant she was named for, had been Oscar Dickson’s partner in funding Nordenskiöld. Sixty-odd years later, one patron’s ship was about to die warning the other patron’s island. She carried six 11-inch guns and a crew of 1,150 ~ it wasn’t enough.

Meendsen-Bohlken wanted the Alexander Sibiryakov intact. Her codebooks and ice reports would tell him where the convoys were. He ordered her – presumably in signals of one syllable ~ to strike her flag, and Captain Anatoli Kacharava gave his answer with a plain-speaking message from his deck guns. Not the best idea, when face to face with the Admiral Scheer, which promptly responded with the few salvoes needed to set the Alexander Sibiryakov on fire. Her crew opened the seacocks so that nothing useful could be salvaged, and most of the 104 people aboard died – but not before they had radioed the nearest military base, with a report on what was happening, and where. Within hours every Soviet ship and station in the Kara Sea knew that a German raider was loose, and the busiest transmitter of all was at Dikson.

BREAKING OUT THE BIG GUNS

With surprise gone and the convoys out of reach, the Admiral Scheer turned south for Dikson. With a radio station, a coal depot, and offices full of ice data – explaining where a ship might sail its cargo of military supplies to meet men travelling overland – Dikson was the headquarters of the western end of the sea route. Meendsen-Bohlken readied a landing party of 180 men -- his intelligence said the garrison was manned by no more than sixty or so frontier guards, bored and bearing little in the way of arms or a credible defence.

His intelligence was nearly right. Dikson’s coastal batteries had just been dismantled for transfer to Novaya Zemlya. Two elderly 152 mm siege guns were sitting on the pier waiting to be loaded. When the warning came, their ammunition was hauled back off the barge and about 300 militia turned out with rifles and hunting guns.

Lookouts sighted the Admiral Scheer at 1.05 a.m. on 27 August. An icebreaker, the Semyon Dezhnev, steamed out to meet her but was holed within minutes, with seven of her crew killed. The steamer Revolutsioner was hit three times and left burning. A third ship, the Kara, sat behind the smoke with several hundred tons of explosives in her holds and was never touched — but then the pier opened fire.

The two guns had no proper emplacements, so each shot sent them rolling backwards. A lorry had to drag them into position again until someone wedged the trails with logs. Crews could barely see their target through the smoke, and as far as anyone can prove, they hit absolutely nothing. What they did do though, was put tall columns of water beside a ship whose captain had been told to expect rifles and relatively little in the way of a defence. The words, ‘cor blimey’ don’t cover it.

Meendsen-Bohlken made smoke and withdrew, circling the island, shelling Dikson’s radio station, its power plant, and the fuel stores. By the time he broke off, he’d used about a fifth of his ammunition ~ his landing party never left the ship.

Admiral Scheer was back in Norway by 30 August, her captain believing he’d wrecked the port. Both damaged Soviet ships were back in service within a week. Meendsen-Bohlken had steered for Dikson with a picture in his head of sixty men with rifles, and that picture was about as current as the spelling on that first map.

It’s almost an addendum, but...

I’m reminded that in his book, Farthest North, Fridtjof Nansen sailed past Dikson on his way east, and pointed out the Taymyr coast disagreed with every map he had.

This second map shows how thinly the northermost edge of the Soviet Union was charted at the time. Those are awfully straight lines. Look:

It’s true, the Kriegsmarine had better material than most, including photographs taken from the airship Graf Zeppelin on her polar flight of 1931. But that still left a heavy cruiser feeling her way along a coastline that bore little resemblance to reality, and between rocks not on any chart. Let’s go back to the first map.

Confident lines suggest a sea-margin well mapped. Bold lines also suggest a country well served from the air, right up to its northern edge. At Dikson, in 1942, that line ended at a barely serviceable airstrip, and little more than a wireless mast, and a coal heap. One starts to wonder, what kind of desperate conversations, maps and charts prompted Unternehmen WUNDERLAND at all.

The prose has not aged well. But between the pages is a hand-drawn sketchmap, and the moment I saw it, I knew where everything was. Simples. That set me thinking, because I also have military maps for the region, which could hardly be less alike.

THE MILITARY MAPPING

This map is what an American officer would have unfolded, come rain or more rain, in the south of France in August 1944. The margin tells us it was “prepared under the direction of the Chief of Engineers, U.S. Army, by the Army Map Service, Washington, D.C., 1944”, and it was “for use by War and Navy Department Agencies only”. That’s always amused me, as I cannot for the life of me image Il Partisano or Elle Maquisard meekly re-foloding any map sheet and handing it over, should they ever find one by accident.

Heights are in metres. A French Lambert Zone III grid is overprinted in red so that a gunner could work out useful references. The roads, also in red, were classified from Michelin road maps, which is a pleasing admission of fault: a husband and wife with a picnic basket and a charabanc would have been less likely to get lost than an army.

Underneath the red, though, the Americans drew almost nothing. The margin says the base was French topographic mapping at a scale of 1:80,000 and 1:50,000, and it names sheet 198 as its source, each one of four quarters dated 1941. That is the Carte d’État-Major, the French staff map, enlarged ~ a nineteenth-century monument to military mapmaking if ever there was on. The Dépôt de la Guerre began trials in 1818; army officers surveyed the nation at a scale of 1:40,000, and the results were engraved on copper at 1:80,000 in 273 sheets. France is not a small nation. As you might imagine, that took a while. The entirety of the country wasn’t fully published until 1875, when the first attempt to show its relief provided hachures to make the point: short strokes running straight down a slope, packed closer and heavier as the ground steepened. Scattered among them were spot heights, single figures in metres fixed by surveyors on summits, cols and crossroads – so the hills east of Montélimar appear as dark, furry caterpillars, and the Rhône plain is just a piece of blank paper.

Hachures do give a vivid impression of shape. What they cannot provide is confidence between spot heights, which is something of a disadvantage to tank commanders.

You can see that a slope on tomorrow’s line of attack is steep; you cannot read off how steep it will be, when – or if – you tackle it. Contours are a better solution, and by 1944, they were, thankfully, almost a standard addition to most new maps. Yet here, in the year of the invasion, and pretty much a product for the purpose therein, the maps issued for that fight still carried the hachures of a century before. There’d been no time to resurvey the Drôme, so Washington reprinted what Paris had engraved, and London had cheerily signed off.

THE SKETCHMAP

Now the other map, the sketchmap inside the book. It has no scale, no grid, no north point, not a single hill, and this is NOT a comparison - no serious tank commander would have depended on such a thing, though more than one officer might have found it useful to help explain his thinking.

The Rhône is a braid of pen strokes down the left edge. Roads are single wobbling lines and villages are small circles with hand-lettered names. Two rivers are named, the Roubion and the Jabron, while simple arrows at the edges point away, and off, to Gap. The artist couldn’t resist adding some decoration, but that’s what artists do. Battlements and turrets run along the top, with the town’s arms between them, an orb topped by a cross. Along the foot is the Montélimar skyline, the castle and its square tower above the rooftops. It is as much a bookplate as it is a map; it’s an echo of war, not a tool for it ~ and yet, curiously, it’s brilliant. Or I think so, at least.

MONTELIMAR is written three times larger than anything else, so the eye is drawn there. Names fall into place: Loriol and Crest across the top, Dieulefit and Bourdeaux to the east, Grignan, Valréas and Nyons along the bottom. Just south of the town are Allan, Malataverne, Châteauneuf and Aiguebelle. Up-river are Savasse, La Coucourde and Saulce. A small flag north of Marsanne marks the 11th Company of the FFI, the French Forces of the Interior -- the sketch puts the whole theatre on a page, which is far, far, more helpful to anyone trying to understand the locations within this terrain.

The odd photographs in this book fill in many blanks. Their captions, translated:

• 5. A heavy artillery piece on rails near Saulce (two are in Montélimar station, four at Livron). Photograph by Dr Cochinard, army medical captain at Valence.

• 6. One of the many Tiger tanks abandoned by the Germans or destroyed by the Americans. Photograph by Dr Cochinard.

• 7. On the airfield at Ancône, near Montélimar, one of the Junkers 88s destroyed by American “Lightning” fighters on 15 August 1944. Photograph by Dr Cochinard.

A small footnote to number 6: to my eye, that’s a turretless assault gun, not a Tiger tank. But then, by 1945, every German tank was a Tiger to those who’d watched them pass; and every dark and pistol-pocketed long-coat was worn by an SS officer, whether he was a card-carrying member of that establishment or not.

SO WHAT HAPPENED, WHERE?

The sketchmap is the echo of a topographic base map, ululating through the summer of 1944. Long before the Allies landed anywhere in Provence ~ Operation DRAGOON, that August ~ local FFI companies and maquis groups were already harassing German forces and punishing collaborators along the strategic Rhône Valley corridor.

Put your finger, as I did, on the sketch just to the south of Montélimar, and one of the worst, best-known events in this region can be followed place by place.

On 30 March 1944, just before dawn, men of the 103rd FFI group were gathered at the railway station of Châteauneuf-du-Rhône, drawing petrol from a tank wagon. They’d been camped for about three weeks in two small farms, nestled on the wooded hill-side between Allan and the abbey of Aiguebelle: one called Nazareth ~ belonging, not surprisingly, to the monks ~ the other, the long-abandoned Laurent farm at l’Aubagne.

A car passed them, travelling almost silently with its lights out. It slowed a little, then turned, and returned towards them. The automatic weapons opened up, and the men of the 103rd FFI group had nowhere to hide. One volunteer silenced the attackers with grenades; Lieutenant Marcel Delaby was killed and Lieutenant Daniel Quineaud, badly wounded, was captured. Survivors fell back to the two farms. The Waffen-SS ~ these men were Waffen-SS ~ followed them up the hill, where the young men held out until their ammunition was gone.

That afternoon, the people of Allan watched those men as they were marched in to the village, chained, one to the next. The book, On Se Bat a Montelimar, explains how they were lined up against a wall at the village school, and the mayor, his deputies and a local constable were fetched to act as witnesses. The parish priest was away, so the Abbé d’Yvoire was called in from Espeluche. He spoke with each man, took their last messages, and then offered himself to the SS commander in their place. The officer declined, thanked him, and made a short speech regretting the necessity. The men were shot in threes, in the back. Vernin lists ten dead from the day, most of them from Lyon, one from Montélimar. He has the wounded Quineaud killed at the same wall, though the Musée de la Résistance en ligne places his death at Viviers, across the river. All ten are honoured, though, in the cemetery at Allan.

Châteauneuf, Allan, Aiguebelle, Espeluche ~ four circles on the sketch, no more than a thumb’s width apart, and the whole sequence of events becomes clear.

AUGUST 1944: THE BATTLE SQUARE

Now move your finger to the north. On 15 August, the Allies landed in Provence, and that same day, according to the book’s anecdotes at least, American Lightnings shot up the Junkers on the airfield at Ancône. The German 19th Army, under General Wiese, started pulling back, up the Rhône ~ making use of the only good road in the region, the N7, which squeezes between river and hills north of Montélimar.

General Truscott of the US VI Corps sent a fast column through the mountains, with the aim of cutting off that direct line of retreat: Task Force BUTLER. On the evening of 20th August, Butler was ordered to seize the high ground just north of the town. By the 21st his men were at Crest and on the ridges above the river ~ the US Army’s official history calls the ground a ‘battle square’: the Drôme to the north, Rhône to the west, Roubion to the south, about 250 square miles. Look at the sketchmap and there it is, drawn by a Frenchman who’d never heard the phrase. Loriol and Crest along the top, the Roubion runs across the middle, and Marsanne, Sauzet, Condillac and Puy-St-Martin lie inside.

The fighting ran for a week, and the 36th "Texas" Division under General Dahlquist came up to reinforce Butler. Against them General von Wietersheim’s 11th Panzer Division fought to keep the corridor open. On the 22nd the Germans briefly took Puy-St-Martin and cut the American supply road. On the 24th they captured a copy of Dahlquist’s plan. On the evening of the 25th American infantry came down from Condillac and blocked the N7 at La Coucourde. Around midnight Wietersheim led a scratch force in person and broke the block. The Americans never did close the road. They shelled it instead. The 36th Division’s own account records more than 75,000 rounds fired, and claims 11,000 German casualties, 2,100 vehicles and 1,500 horses, with wreckage for sixteen miles. It also counts six 380 mm railway guns. Dr Cochinard’s caption counts them too: two at Montélimar station, four at Livron. Montélimar itself was free on 28 August. Much of the Nineteenth Army got away across the Drôme, but in doing so, it also left much of its transport behind – not needing any map, never mind an accurate one.

WHAT A SKETCHMAP CAN DO

The official map sheet tries to answer questions a soldier might ask. How steep is this hill, where is that bridge, what’s my grid reference? It is a remarkable object, showing the work of French surveyors who passe that way, a century before.

But to me, the sketchmap better answers the question that a stranger might ask first: what happened, where? Someone drew only what mattered to the story and left the rest of the Drôme as white paper. That choosing is a skill: it is useful to know what the task in hand is, and a map doesn’t always need to be detailed, to be helpful.

I’m still not sure which I prefer, the sketchmap or the map sheet. Both are on my desk, and I shall put the book back in its rightful, new place on the shelf ~ eventually.

Maps are a weapon – a powerful weapon – because maps look, simply, like facts. A map says, ‘this is how things are’, and a mapmaker can change the world... so it’s no surprise, mapping had its run as a propaganda tool during the Second World War.

Our main map in this article, for example, begs the question: is it a propaganda map of Portugal or a Portuguese propaganda map? The two are very different, and between 1939 and 1945 the difference mattered. A Portuguese propaganda map was one made by Portugal, for Portuguese purposes. A propaganda map of Portugal – or, more often, one dressed up in Portuguese – was made by someone else entirely, and slipped under Lisbon’s door. Portugal was neutral, you see, which made it one of the few places left in Europe where both sides could still advertise their intentions: British and German propagandists worked the same streets targeting the same people, each assuring the Portuguese that the other lot were dastardly fellows.

A map was ideal for this. There’s a snag though, in this propaganda metier, and it’s worth keeping at the back of our minds as we admire the pretty pictures; no matter how good the mapmaking, the actual map can only informa and persuade a person who’s able to read it.

THIS IS NOT A SMALL COUNTRY

Let’s start with the real thing. Portugal não é um país pequeno – ‘Portugal is not a small country’ – was produced in 1934 by Henrique Galvão, an army officer and colonial administrator, for the First Portuguese Colonial Exhibition in Porto; an exhibition he also happened to be directing, which does rather simplify the commissioning process somewhat. The map was printed by Litografia Nacional, of Porto, and uses a simple trick that’s mirrored today by online whizzymajigs. thetruesize.com is one such.

Take Angola, Mozambique and the rest of the empire; colour them brightly; and lay them down, right across Europe, as if you’re floofing a blanket. Add a table of figures proving that Portugal and its colonies, together, outsize Spain, France, England, Italy and Germany combined. Small country? Certainly not, mon ami. Just badly folded.

This map worked so well, Salazar’s regime reprinted it for decades: postcards in Portuguese for home consumption; French and English versions for visitors; and - by 1951 - an edition from the Secretariado da Propaganda Nacional laid the whole lot over the United States instead – for the benefit of American tourists, who presumably needed a unit of measurement they could relate to. People ask where Brazil has gone. Independent since 1822, I’m afraid – a century too late to help with the arithmetic.

Galvão himself deserves more than a footnote. Post-war, in 1947, he reported on forced labour in those same colonies, concluding only the dead were truly exempt; he fell out with Salazar, was arrested in 1952, escaped – and in 1961 hijacked a liner, the Santa Maria, to make his point. Not many mapmakers manage a second act like that.

THE MAP NOBODY SIGNED

Back to the map at the top. Mappa de Europa carries no date, no artist, no publisher and no place of printing – which is exactly what you’d expect from someone proud of their work. Its message is anything but subtle. The Soviet Union is a ragged, barefoot brute – burning churches, torching farms, enslaving Ukraine and the Baltic states – and its only friend is Britain, drawn as a stout and faintly sinister gentleman whose ships are busy sinking.

Against them stands Germany: clean-shaven, well-dressed, and joined by soldiers from Spain, France, Belgium, the Netherlands, Denmark, Norway, Finland, Italy, Czechoslovakia, Hungary and Romania. Those, for the record, were volunteer brigades of the SS rather than the defeated governments of half a continent – but the map isn’t minded to explain that. And Portugal? Portugal’s soldier has turned his back on the whole business, and is gazing out at the Atlantic. The events shown put the date at about 1942.

There are two schools of thought around this map’s origins. One says it appeared first in Lisbon: this is Salazar’s balancing act on paper, protecting the ancient alliance with England without cosying up to Berlin. The other says it was drawn in printed in Berlin, as a nudge for the Portuguese to stop staring out to sea and join in on a merry crusade. So, who made this map?

For me, it’s the spelling that makes things clearer. No native Portuguese speaker in 1942 would have written mappa with a double ‘p’ – Portugal’s spelling reform of 1911 had seen to that. Accents are missing from Italia, Eslovaquia and Paises Balticos. Hungary has lost its H altogether to become ‘Ungria’, and Denmark appears as ‘Danemarka’, which is how nobody in Lisbon – or Rio, come to that – has ever spelled it, no matter how many pints of Porto they’ve consumed.

In fairness, a Brazilian national would have written Mappa da Europa, not de Europa, so the grammar is Portuguese even where the spelling isn’t. It’s either a Portuguese map with oddly archaic spelling, or a foreigner’s map with a phrasebook.

But then there’s the drawing, itself. The illustrative style is pure Franco-Belgian comic strip – very much à la Tintin – and Hergé himself spent the occupation drawing for a German-controlled newspaper in Brussels; he wasn’t the only illustrator of that school to do so. Add the label ‘Greater Germany’ – the Nazis’ own name for their enlarged Reich – and this looks rather less like a Portuguese map, than a map wearing a Portuguese hat.

IS ENGLAND IN TROUBLE? YES! (DO MAKE UP YOUR OWN MIND)

The second anonymous offering is less coy about what’s what, and which side it’s on. Encontra-se a Inglaterra em apuros? Sim! Que cada qual forme o seu juízo! translates as ‘Is England in trouble? Yes! Let everyone make his own decision!’ – a question, followed by a nuance answer, from a sheet that’s already made up its mind. Britain sits in the middle, almost hidden under bomb damage: London, Southampton, Plymouth, Cardiff, Bristol, Coventry, Norwich and Kingston-upon-Hull. Also known as ‘Ull.

Around the edge run six panels dressed up as frames from a reel of film – British ships going down to U-boats; to aircraft; to surface raiders – and two long quotations describing the general state of affairs in Britain: food running out, a government failing and citizens in a mad, mass panic. Those views are credited to the Sydney Sun, quoting The Times of London... a chain of custody that ought to trouble any editor.

On the date? Experts disagree. Cornell University has this map stored with a label saying printed in 1941, and suspects Nazi sympathisers working inside Portugal; The Map House in London has it appearing more loosely in 1943, and blames the Axis propaganda machine directly. What nobody suggests, is that Salazar’s government printed it. This is a map of Britain, written in Portuguese, made by Germany’s friends – not a Portuguese propaganda map, nor even a propaganda map of Portugal, but a propaganda map for Portugal.

Do keep up.

LONDON REPLIES, LISBON LAUGHS

The British weren’t going to let this stand. In 1943 the Ministry of Information issued The Battle of the Atlantic: a pictorial map by Frederick Donald Blake – Greenock-born, Camberwell-trained, and also found drawing war maps for the Daily Express. Convoys steam east under escort; bombers pound the U-boat pens at Lorient and St Nazaire; everything is going splendidly. It was printed in English, French, Dutch, Arabic – and Portuguese. Same ocean as the German sheet, with the same ships, but a rather different message. Meanwhile, the Portuguese were drawing maps of their own, and these are the ones I like best.

António Soares – a modernist painter who took a Grand Prix at the Paris Exposition of 1937 – dusted off an animal map he’d drawn for the First World War, and updated it for Livraria Franco, a Lisbon bookseller, in 1939. Germany is a tiger; Russia a polar bear; France a lion; Switzerland a lamb. Portugal is a white dog baring its teeth – a bold little bit of anthropomorphisation, for a neutral country.

Carlos Ribeiro’s map of 1940, for Livraria Barateira, is far less flattering. Germany is an octopus, throttling Poland. Switzerland is a dove in a cage. Portugal is a snail.

Neither is a map purporting to be state propaganda – they were sold in bookshops, to people who fancied a laugh. But they are Portuguese maps, not afraid to state their makers’ personal points of view, and that’s more than can be said for the anonymous items being printed elsewhere.

YES, BUT DID ANY OF THESE MAPS WORK?

The troubleis, to be moved ~ or influenced, at least ~ by Mappa de Europa, you must first know which blob is Ukraine, where the Baltic is, and that the small rectangle on the far left with its back turned is you. In the early 1940s, that’s still asking a good deal of Joe Public, Zé Povinho, or indeed, José Público.

Portugal’s census of 1930 classed roughly six of its own people in ten as illiterate. By the early 1940s, the regime was celebrating the fact that – for the first time in the country’s history – a bare majority could read and write. Reading a map is something else. Plenty of people who could manage a newspaper headline still couldn’t draw an outline of Europe, let alone pick out Kingston upon ‘Ull.

I suspect the propagandists knew this – which is why these maps are barely maps at all. Take the geography away and the pictures still work: the barefoot brute is easy to understand, as is the burning church, the sinking ship, and the soldier with the good jaw. The map is there to lend authority; the cartoon does the persuading. You don’t need to find Coventry on a map, to understand a drawing of it on fire.

And the result? Well, as we now know, Portugal stayed neutral right ‘til the end. It granted the Allies permission to set up a base in the Azores in 1943 ~ was rightly miffed about what happened on Portuguese Timor in 1942, and THAT’S an article still to come ~ and joined nobody else’s crusade in the interim.

Whether any of that owes a thing to a film-strip of sinking freighters, or a snail drawn for a Lisbon bookshop, is doubtful. Salazar read the other sort of map – the sort that showed where the armies actually were – and drew his own conclusions. Which rather leaves these propaganda maps doing what they do best, eight decades on: looking marvellous on a wall, and telling us rather more about the people who made them than the people they were aimed at.

Zé Público, one suspects, peered at these maps and went back to his pastel de nata.

All credit due, as always, to the purveyors of fine mapping such as this ~ including Barry Ruderman, Curtis Wright, Geographicus, et al.

This week, the chaps at We Have Ways gave us a top-notch gallop through the Allies’ advance to Ortona, on the east coast of Italy. Few can match when it comes to finding personal stories that bring history to life, and this small corner of the Second World War is a dark moment that merits more time in the spotlight.

Even so, through no fault of their own, both James and Al Murray missed one crucial aspect of those events in November and December 1943. Much of the Allies’ hardship and most of the tragedy can be blamed on the maps. Here then, is a hasty follow-up with an overview on some of that mapping, and the reasons it cost so many lives.

ORTONA IS ~ WHERE, EXACTLY?

Let’s start with a little geography. We tend to think of Italy as a north–south affair; a thin boot sitting tall, garter to stiletto heel. Histories of the Allied thrust offer a tidy picture, giving the impression of a slog ‘north’ towards central Europe. Men climbed the leg inside that boot: Americans on the left, the British and everyone else – Indians, Canadians, New Zealanders et al – on the right, all aiming to meet in Rome. Geography is a strange thing. So is history. Neither pays much heed to tidy pictures.

In reality, the boot sits at a slant: the defensive lines constructed by Organisation Todt ran south-west to north-east, as you can see above ~ at right angles to the direction of travel, giving short quarter to the terrain. Ridges, rivers – pulsing varicose-veins, such as the Sangro, and the Moro – as well as deep, broad, U-shaped ravines, like the one that the Canadians referred to as ‘the Gully’. That, formed by a creek called the Fosso Saraceni, lay just to the north of the River Moro, south of Ortona town proper. The Gully had to be crossed. There was no going round it.

In the winter of 1943, the Abruzzo was a brutal region. Picturesque, but brutal. Passes and tracks and switchbacks; acacia bushes; splintered vines; rolling, spine-pricked waves of verdant undergrowth; soap-scented olive groves; ruinous rural settlements and shattered lives; and the rotting corpses of bloated pigs and dead pack-mules. Men, too. Butchered and blackened, marking the progress mile by mile, of armies suffering badly in a harsh, wet winter. Ortona itself – a medieval, deep-water port, essential to supply lines – sits on a headland: a strategic checkpoint; a coastal chokepoint; and – in the 1940s – poorly served by the few metalled roads running in and out of it. Tight, tall and narrow at its heart; dense alleyways that, over time, had expanded into neat, criss-cross rows of modern construction ~ the new town to the south-west.

In a bird’s eye view, those streets were logically defendable, depending on who was trying to defend them. And to be honest, the answer to ‘Will a Sherman fit down that street?’ is usually, ‘It will in a minute.’ This is the town at a large scale.

Scale is important in this story.

A ONE-MINUTE MAP LESSON

Scale. A map at a scale of 1:50,000 shrinks the world 50,000 times. One centimetre on paper is half a kilometre on the ground in metric terms, so a field and a farmyard are a few millimetres apart. At a scale of 1:25,000, everything is twice the size: four centimetres to the kilometre, or about 2½ inches to the mile. That scale shows individual buildings, hedges and tracks – it’s the fighting scale for most armies.

After the First World War, the British Army decided 1:25,000 was the better option for close conflict, and the ideal solution for its artillery. Ordnance Survey started – slowly – creating a 1:25,000 map of Britain, tactfully calling it a ‘2½-inch map’ so that nobody would notice the change to metric measures. Post-war budgets were tight. It paid little to infer too much progress, too far, too fast. But by 1939, most of the British Army had agreed which maps it wanted, wherever it was being sent. Having those maps in hand, and being able to trust them, was another thing altogether.

The grid. Military maps carry a grid of numbered squares. Any point within them can be given as a 6- or even an 8-figure grid reference, which works much like a postcode for a patch of ground, some 100 metres deep and wide. A gunner needs two of these references: one that defines where his gun is accurately, and one that has a stab at pin-pointing the target. From those numbers, he can calculate a direction and a distance – squint a bit, think about parabolas – set his sights, and fire.

Gunners do not need to see their targets. That is the joy of geodesy: the science of accurately measuring and understanding the Earth’s geometric shape, gravity field, and orientation in space, and how those properties change over time – it’s the very thing that gives men a quasi-religious faith in mathematics, because as long as the maps and surveys are accurate, it delivers repeatable miracles. Remember that.

WHERE ITALY’S MAPS CAME FROM

Before 1861, mainland Italy was a patchwork of kingdoms, duchies, republics and papal states, each with its own surveyors, measuring units, and opinions about how a map ought to look. Teams hauled survey instruments up mountain-sides to fix survey points for posterity: stone pillars on hilltops, iron bolts sunk into bedrock and church steps, wooden signal frames on the summits. It was back-breaking work, done to the best standards of the day, and Italian mapmaking was – still is – admired far and wide. This is the land of explorers, after all.

If we set aside the truism that pack-mules occasionally decided this was close enough to that summit – then, all in all, the results were actually pretty good, bearing in mind the time, the budgets, and the technologies available. It’s what happened next that matters. The trouble, in 1943, wasn’t the quality of those original surveys. Rather, it was the cultural affinity with ‘domani, domani’ that had most impact, along with the economy, and the politics of the day. Large scale maps of the plains and industrial north were revised ~ keeping up with changing appearances ~ while the agricultural south and undulating interior, simply weren’t. And at scales of use to the military, such as at 1:25,000, any sheets that were printed in the interim, were usually made with contours inked in black or brown alone. In hill country, Italian mapping looked like fingerprint whorls. Prime example, voila:

That was Italy’s military mapping legacy, shared far and wide. It was not unusual for countries to swap / acquire / purloin / purchase entire map series of other nations. Now add a war, light the blue touchpaper and stand well back.

HOW THE ALLIES MADE THEIR MAPS OF ITALY

The Allies – the British, the Canadians, the Americans et al – did not send field survey units to double-check or update Italian counterparts’ work of ages past. There wasn’t time. There weren’t the spare resources. There was no real opportunity either – there was a war on, dammit.

Instead, Britain’s GSGS – the Geographic Section of the General Staff – photographed and produced copies of maps that had long-since been made by the Istituto Geografico Militare, the Italian army’s mapping service. Reprints of old maps, based on geriatric surveys. The first British edition of the 1:50,000 series (GSGS 4229, if you’re keeping score) was a black facsimile of the Italian map, made at the same scale. The only addition was the South Italian Grid – in other words, the British took an Italian map, printed a British grid over it, and called it a British map. Complete with its original errors; all beautifully photographed, and very slightly blurred.

Later editions added colour. Detail was printed in a grey-blue ink – gris-bleu in French – which British soldiers promptly called ‘griblets’. The colour changed to brown, the name stuck. Roads were picked out in red; water in blue; a diagram in the margin showed which parts were ‘revised from air photos’ – spoiler alert, not very many – and apart from that, the word ‘revision’ did the kind of heavy lifting usually associated with body-builders. The paperwork said maps had been revised. The body-count soon started casting aspersions on that statement. Still though, the real problem was the age of the originals. The north of Italy, Sardinia, and the west coast from Naples to Rome had been surveyed from around 1930 onwards. For the rest of the country, most maps were pre-1920, many of them still reproducing the detail from surveys done in the 1860s. Just think about that for a moment. Really think about it.

A nation coming together; investing; constructing roads, albeit slowly; developing its infrastructure as fast as it could – was defended and attacked with maps drawn from views last seen 70 years earlier. In December 1943, Canadians fought up the Adriatic side of the mainland across 20th-century ground using maps drawn for 19th-century farmers. Some of those farmers’ grandchildren were probably shooting at them.

THE 1:25,000 AND 1:250,000 MAPS THAT WEREN’T

The artillery expected maps at a scale of 1:25,000. In theory, that’s easy. Take a map at a scale of 1:50,000 and enlarge it to twice the size. Voila. And that – where no decent Italian 1:25,000 map was to hand – is exactly, unfortunately, what the Allies did.

Enlarging a map is like enlarging a holiday snap of an old flame. Everyone gets bigger. Nobody gets any better-looking. A road that is 200 metres out of place on a small map is still 200 metres out on the large one; it’s just more convincing and easier to believe. Proper, new 1:25,000 sheets for the Italian mainland – updating old maps with details gleaned from recent aerial surveys – only started to appear for the Allies in the spring of 1944. Ortona had to be taken in the winter of 1943. Whether the sheet on a Canadian gunner’s table that December was an Italian original or a blown-up 1:50,000, it still rested on the same elderly survey. It looked like a 1:25,000 map. It wasn’t a 1:25,000 map in any sense that mattered to a gunner. And what’s just as relevant, is the maps used in planning that gunner’s action – collation maps at 1:250,000 scale.

This excerpt is from one such sheet, printed by 512 Field Survey Company RE, made by ‘Staff Officer (Intelligence) Levant’ in April 1943. The chances of it being updated before the Canadians entered the scene were remote. Maps at this scale show rivers, roads, rivers, and major towns that – unless it’s a map of Coventry or Dresden – are unlikely to change much over the space of a few months. Caveats do abound though, a few of which are evident here:

On any collation map though, general topographic features are the main event. Seeing them at this scale, strategic decision-making – projected lines of advance; assumed lines of objection; expected lines of conflict – become understandable.

At this point, James Holland and Al Murray remind us this was one of Italy’s harshest winters; rivers were torrents; the mud and the blood were intermingled; the Allies had pushed forward with pack-mules supporting stretched supply lines; Germany was under pressure but that pressure was exerted on well-established defensive positions; and Abruzzo – beautiful, scenic, breath-taking Abruzzo – was a region riddled with sky-scraping ridges, thorn-filled ravines, craters, shale slopes... you get the idea.

“REVISED IN OCTOBER…”

Canada’s post-mortem on this front is bleak by any standard: “... during January 1944 – 1944 – Eighth Army fought along the Ortona–Guardiagrele front where the 1/50,000 maps were found to be very inaccurate. They had been revised in October from incomplete, small scale photo-cover, but it was clear the original survey was of inferior quality. More accurate and legible maps were requested by the artillery.” One can only imagine the tone of voice around ‘requested’. The phrase ‘small scale photo-cover’ is doing a hell of a lot of lifting.

To update a map with details from the air, ideally those details are gleaned by the right aircraft, carrying the right camera, flying steadily and predictably in a pattern that runs to-and-fro, and always at exactly the right height. Not almost. Exactly. The ideal bird for the task was the Mosquito; fast, twin-engined, with room for more than one camera. The RAF wanted its Mosquitos for reconnaissance and was extremely reluctant to release them for survey work – understatement of the war.

Three were sent to the Mediterranean. Those were recalled. That left survey units largely dependent on the American F-4, a photographic version of the P-38 Lightning. To stay alive, it flew at 25,000 to 30,000 feet. From that height its standard 6-inch camera produced photographs at roughly 1:50,000 to 1:60,000 – which was about the same scale as the very same maps that needed correcting to 1:25,000. Imagine it this way: try to update a street map from a photograph the size of a postage stamp. A new coast road might just show up; a new farm track, a gully’s exact edge or a farmhouse’s true position – would not.

So the October 1943 revision didn’t just enlarge old, bad maps. It also tried to update some of them by referring to photographs that were too small and didn’t cover everything. The result? A very bad old map with a more recent date on it.

WHAT WAS ACTUALLY IN THE MAP CASE

Two items printed during this period survive as originals: a 1:50,000 defence overprint of the countryside around Ortona, and a GSGS 4380 town plan at 1:5,000, titled Porto di Ortona. The defence overprint was a griblet, with known and suspected German positions printed on it in a second colour – all of which had been interpreted from air photographs. Depending on which one you held, the same farmhouse could have two or even three different grid references and, crucially, could have 500 yards of error baked into its margins.

As to the town plan? Well, GSGS had prepared some town plans for the Italian mainland, long before landing on it – but predictably, smaller coastal towns were drawn from old, Italian harbour charts. Corrections from aerial surveys were hit and miss – mostly miss – and this is as good a point as any, to point out this was a world-wide war, with only so many mapping resources. Town plans for small, coastal Italian towns were not high on anyone’s agenda. Besides, Ortona was a town under attack, and few plans survive that for long.

From the air, the Royal Canadian Air Force’s No. 417 Squadron; the RAF, and the South African Air Force all had at it. From the sea, the Royal Navy’s hurled 4.7-inch guns fired high-explosive shells weighing roughly 23 kg each, 10 to 12 rounds per minute per gun, onto the castello used by German observers as a primary artillery spotting tower, and into the northern outskirts of the town, targeting the coastal highways and secondary roads. On the ground, the attack into and through the town itself involved M4 Shermans of the Three Rivers Regiment; QF 6-pounders; No. 75 Hawkins Grenades & Teller Mines; Pak 50mm & Pak 40 75mm anti-tank Guns; turretless StuG IIIs ...

... and in that kind of attack, any town plan loses relevance quickly. But as we now know, the trouble started long before the Allies reached Ortona.

Some roads on the maps available – weren’t roads. For a start, while the road through S. Apollinare and S. Leonardo had been designated as the main Corps axis, a newly-constructed road from S. Vito Marina to Ortona followed the line of the coast more closely, but its completion was too recent to be shown on the maps available.

Moving up, as patrols neared the Moro’s south bank, three other routes had looked like good choices: a half-built highway near the river mouth; an old mule path passing Sant’Apollinare, dropping into the valley; a rutted inland track skirting farms around Rogatti. Each had pinched progress, with bad going reducing movement to guesswork. Still, that crossing had been forced, and San Leonardo was taken. The way north ought to have opened up then, towards the Cider Crossroads. Instead the land had simple fallen away into the Gully, and every push in any direction had found the same problem: German paratroopers, mines in vineyard rows, and machine guns set where they could rake the approaches. The weather had closed in, too. None of this helped.

What’s more, British grids offering references to mark a lack of progress didn’t agree from one map to the next. Rivers ran 100 metres from where they should and the rain – oh heavens, yes, the rain in Italy in the winter of 1943 – turned those 50-yard rivers into 300-yard wide torrents. The same roadside inn appeared again and again as ‘Osteria’ – not everyone spoke Italian and mis-transliteration was common - and a village called Villa Rogatti was Villa Roatti, or – on German maps pillaged en route – Villa Ruatti. Few editions agreed with each other. They had all been copied, badly.

At 1st Canadian Division HQ, sheets at 1:100,000 were doing the rounds – Sheet 141 of the GSGS 4164 series, CHIETI, showing Ortona town, and Sheet 147 – LANCIANO – showing the Moro, San Leonardo, and San Vito, looking as though they’d both been found in the bottom drawer of a very old library, not without cause. Finer details of the Gully weren’t on those sheets. If you’d been the officer examining Sheet 141-II-SE at 1:25,000 – ORTONA, from GSGS 4228 – or the Ortona 1:50,000 sheet from GSGS 4229, 141-II ... then you’d have discovered the boundary between them sits at exactly 42°20′ north. Ortona is about 2½ km north of that line. The Moro, San Leonardo and San Vito are south of it. The Gully and the lateral road lie roughly along the join, so you must use two maps to plan anything – and you just know, don’t you, that those map sheets don’t meet neatly, anywhere.

The Gully – that dip in the ground – was assumed to be a German defensive position. Few realised, from the maps, it would be too steep-sided for tanks, and impossible to get round easily under fire. On anything tactical, it had been reduced to a thin line – more an inconvenience than a decisive point in the landscape. Troops of 90th Panzer Grenadier Division and then 1st Parachute Division held it for the best part of a fortnight: and for the record, Germany’s mapping of Italy was based on – you guessed it – the same, old, Italian sheets at 1:50,000.

This is not the moment to talk about the Italian Military Geographical Institute, and its cooperation with Germany’s Department of War Maps and Surveys. There will be other moments to do that. This example of Italian maps, reproduced by Germany, puts much into perspective:

The Allies’ own sheets misled the Canadians. On any maps to hand, ‘Berardi’ looked like a hamlet – some reports even spoke of ‘street fighting in Berardi’ – but in reality, Casa Berardi was a large farm: a house, a few outbuildings, a dog, his bone, and very little else. Capt. Paul Triquet of the Royal 22e Régiment was awarded the Victoria Cross for his part in the fight for Berardi on 14 December, without the benefit of any maps. This isn’t pedantry. If you think you’re attacking a village, you plan for a village: its time, its troops, its fire. If you arrive to find a farmyard, or if the farmyard turns out to be a fortress, your plan of action is a fiction from start to finish.

RANGING. OR “HOW GUNNERS FIND OUT THE MAP LIED”

This is where bad maps truly stop being an inconvenience and start killing people. A gun doesn’t hit what it’s aimed at the first time, and gunners know that. The remedy is called ranging. Before a bombardment – in an ideal situation – guns fire a few shells at a point safely short of the enemy. An observer sees them land and radios back with adjustments – I’m making no apology here for describing this in simple terms.

Gunners make adjustments. They are good at doing this. Once the shells start landing where the map says they should, that becomes a fixed ‘control’. From there, a barrage can be lifted forward in calculated phases – it is a thing of wonder, when it works. Shells land further and further forwards, mere measured yards at a time, with the infantry following on behind... taking advantage of newly-churned ground as cover, ducking for cover from still-falling shrapnel, but making the most of gun-smoke (acrid) and chaos (acres of it).

Ranging, however, has two costs. It tells the enemy that an attack is imminent, and it needs someone who can see where the shells fall. By 1943 the Royal Artillery was also predicting fire: ideally in a moment of calm, it could calculate every gun’s aim from maps alone and open fire without a single ranging shot. The enemy gets no warning. The infantry get a barrage that is exactly as accurate as the map, and no more – and now consider what that meant, at Ortona.

The survey under the grid on Allied mapping dated from the late nineteenth century. It had never been properly updated since. In places, the large-scale sheet was just the 1:50,000 blown up to twice the size. The grid printed over it was mathematically perfect – but the hills, roads and gullies sitting underneath it all had various opinions about where or where they did not actually exist. As I said, errors of up to 500 metres were noted, here and elsewhere on the Italian mainland. A predicted barrage takes every one of those errors on trust.

Then add the weather. In mid-December, cloud sat low on the approach to Ortona; and anyone who’s spent time in Italy swatting mosquitoes knows that same cloud drops low to kiss the mists rising from any watercourse. Even if someone had wanted to check the fall of shot, observation was unreliable at best. Nobody could see where shells were landing, until they landed. In fairness, guns can drop short for other reasons: worn barrels, cold charges, the wind. Ranging exists to catch all of those too. At Ortona, that ranging checks might have prevented carnage – though again, simply ranging isn’t enough. The maps have to support the action.

MORNING GLORY

On 18 December 1943, the Canadians launched two attacks behind two fire plans, codenamed MORNING GLORY and ORANGE BLOSSOM. Flowers, both.

MORNING GLORY, supporting the 48th Highlanders of Canada, opened at 0800 hours with a deafening roar. It filled the air with the screams and sighing of passing shells and laid a wall of bursting high explosive 1,000 yards long and over 300 yards deep. The Highlanders went forward by compass, with the Sherman tanks of the Three Rivers Regiment grinding along beside them. Picture the gun lines while it went on – firing until the paint curled from the red hot steel and the men reeled in the pits from exhaustion and were sick from the blast – that’s not my imagination, it’s a quote from the historical record.

The Highlanders took their ridge, but the fire plan had already given the game away. What’s more, on the ground, the objective was 300 yards north of where it had been located on the map, which meant protective fire was falling on the objective rather than beyond it. Three hundred yards out, for the first barrage of the morning. The map was wrong, and nobody had fired a shot to find out beforehand.

ORANGE BLOSSOM

Shortly before midday the second barrage opened, carrying the Royal Canadian Regiment forward. In theory, ORANGE BLOSSOM was to have fallen parallel with the lateral road while the infantry used the railway track as an axis of advance. For the second time, the guns of 5 Corps opened fire. While the shells were still bursting on the opening line, 3 Canadian Infantry Brigade reported the rounds were falling short again – this time landing on the Carleton and York Regiment, despite the fact the Battalion had been withdrawn 300 yards from the edge of the gully.

Consider the words, ‘the opening line’. That is the line ranging would have fixed, the first line of the barrage, the one everything else is measured from. This time it was set from the map alone. It was wrong from the first salvo, and every lift forward was calculated from it. ‘Reported the rounds were falling short’ – is a war-diary voice, written for the footnotes of history. What it means is this. The Carleton and York Regiment had done the right thing. They had pulled back 300 yards from the edge of the Gully, the safety margin the plan asked for, and waited in their slit trenches for their own guns to clear the way. Then their own guns landed on them. The shells came in from behind, from the side they had every reason to trust.

A barrage lands as a wall of high explosive, not one shell at a time. Men who’d braced for the enemy found the ground erupting around them, churned up by shells shot by their own side. Through the smoke and flying earth a figure might loom up and vanish again before anyone could put a name to him. There was no cover facing the right way; every direction was the wrong way. Wounded men called for stretcher-bearers who had to cross the same ground to reach them. Officers screamed down dead or dying wireless sets to stop the guns, while the fire plan carried on against its issued timetable, because, dammit, that’s what fire plans are supposed to do.

Men died, and others would carry their injuries and that morning for the rest of their lives. ‘Falling short’ does not describe it. The guns were firing on grid references that didn’t match the ground, with no ranging to show it and no clear sky to see it. The first some knew of the error was when the screaming grew too loud.

LESSONS LEARNED, THE EXPENSIVE WAY

In the end, the Canadians took Ortona. They crossed the Gully, reached the town and cleared it house by house in the week before Christmas, knocking holes through party walls because the streets were death traps. That is a summary in a sentence. Read ‘The Savage Storm: The Battle for Italy, 1943’ by James for more details – I’ve included a list of useful books below, and will try to do that always, going forwards.

The 1st Canadian Infantry Division lost well over two thousand men ~ the press called it Little Stalingrad ~ and the Allied learned many lessons. Some of those included the fact that, still, four years into the war, too few understood how maps were made.

Moving forwards, field units in Italy started producing new 1:25,000 ‘key’ maps for the artillery: outlines of roads, rivers and landmarks, drawn from air photographs. An error of 50 metres was accepted as good enough, and fresh map sheets could be turned out in about four days. By the spring of 1944, at Cassino, gunners were working with brand new 1:25,000 sheets; the German defences was overprinted in red, and artillery targets were overprinted in violet. Some still carried a warning in the margin: NOT TO BE USED FOR ARTILLERY FIXATION. It’s hard to read that without thinking of December 1943.

It is also tempting to file what happened in the advance to Ortona under ‘friction’, the gritty get-out clause, Clausewitz insisted be inserted in every textbook. To my mind, the events here are an example of history getting the better of geography. Historic surveys informed current mapping; map sheets missed roads nobody had thought to draw; rivers and ravines were misunderstood in more than one dimension; place names were added in error; features existed or did not exist where nobody had thought to map or map-out a plan; buildings were dressed as villages, and a grid sat over a survey that was three times the age of men trusting it with their lives.

And you thought the battle for Ortona was all about mouseholing.

• James Holland, The Savage Storm: The Battle for Italy 1943 (2023). This covers the campaign from the Salerno landings in September to the end of Ortona in December. It’s good on the wider Eighth Army picture and the German side. (Follow it up with Italy’s Sorrow about 1944–45.)

• Mark Zuehlke, Ortona: Canada’s Epic World War II Battle (1999, with several later editions). An account built on interviews with veterans. It follows the fighting from the Moro through the Gully to the house-to-house battle in the town.

• Daniel G. Dancocks, The D-Day Dodgers: The Canadians in Italy, 1943–1945 (1991). A lively history of the whole Canadian campaign in Italy, told largely in the soldiers’ own words.

• Farley Mowat, And No Birds Sang (1979). The memoir mentioned by James in this week’s podcast episodes, written by a junior officer serving with the Hastings and Prince Edward Regiment. It covers the Moro and the approach to Ortona, and is one of the great Canadian war memoirs.

• Marco Patricelli, La Stalingrado d’Italia: Ortona 1943, una battaglia dimenticata (UTET, 2002). The main Italian history of the battle, including what it did to the town and its people.

• Andrea Di Marco, Assolutamente resistere (Menabò, 2013). The battle from the German side, drawn from war diaries and soldiers’ testimony. Di Marco also tries to identify the Germans who died there.

• Canadian Military HQ Report No. 165. Do have a look at the Historical Section’s account. It’s free on the Canada.ca site, though you’ll have to download it in parts.

  Finally - here’s the link to those We Have Ways pod episodes.
  If you’re not listening to it, why not?

ISLANDS OF REALISATION

Standing on the quayside in Symi harbour – ice-cream in one hand, squinting at the map sheet that’s flapping in the other – the town itself stands beside you, above you, and behind you. Terracotta façades; burnt-sand pediments; bright-white stucco walls, and hints of periwinkle blue paint – stackable, triangle-topped houses that clutter the harbour’s amphitheatre in tiers; Gialos, the buildings at ground level, Chorio the village above it, and behind them both, the island simply rises up, and up, and up.

The map sheet – AMS M801: Sheet 23, ‘Isola di Simi’, 1:25,000, first printed by the US Army Map – is awesome. One piece of paper shows everything you need to see on Symi: the paperclip-curves of deep-water harbours; tracks across the island, chapels here and there; and terrain tamed in ten-metre contours – it is an unusually good map, both for the time and for a small Aegean island. It explains, quite accurately, that the ground here rises. It does absolutely nothing to prepare you for how.

On the ground, and on the map, the town of Symi is divided from the village of Pedi by a ridge-line. That sloping ground hides one from the other: the road between them does not look troublesome. Red-inked and wriggle-free, it rises modestly and then switches back on itself, almost apologetically; accommodating the contours that give one part of Symi town the nickname ‘Alaska’ – the sun caresses houses briefly on one side – and lets ‘San Francisco’, opposite, bask in a searing, day-long light. In fact, Pedi itself is barely a mile away as the crow flies. But then you look up, and realise even Greek crows might think twice about flying there directly.

It is a surreal situation. Holding a map of Symi in your hands, and understanding how to travel around this island and fight to defend it, are two very different things.

The island’s interior has arranged itself like a crumpled pile of washing thrown down in haste by Greek gods. Gullies appear where the map suggests slopes. Paths look like shortcuts but become climbs over loose shale; heat alters distance, and even that relatively new tarmac road – one of the better stretches of road-surface here – has to find its way across a ridge that has no interest in making movement convenient.

I visited Symi partly because my brother has a mind to move there – who knew,_, Big Sisters who’ve lived abroad might have views on that kind of life-changing event – and partly because of this map. Produced in 1943, the entire Dodecanese series at 1:25,000 was based on Italian mapping made in 1929, with its roads revised by 512 Army Field Survey Company, Royal Engineers – mainly from air-reconnaissance, mainly done by bombers with little training in flying straight and level repeatedly with specialist, cameras. That said, it is a good map. But again, its contours don’t tell the whole story.

As soon as you consider going from one side of Symi to the other, you face a problem. For centuries, the easiest way to make that journey has been to sail around the island. In September 1943, a small group of British soldiers faced the same problem in far less agreeable circumstances – and without the benefit of ice cream and water-taxis. Those men had access to this same map. What they needed, was knowledge of the ground.

WHAT IS WHERE

Lying to the north-west of Rhodes, Symi is a 66 square kilometre barnacle, with its highest point – Vigla, meaning ‘lookout’ in Greek – at 617 metres above sea-level. The interior is dominated by steep-sided, rigor-mortis-grey limestone reaches; the coast is broken by nip-tuck bays and jagged coves; and land-lights on the Turkish headland are visible at dusk – your roaming charges get argued at a later date.

If you believe Homer, then Agamemnon’s fleet included three ships that sailed from Symi to besiege Troy. The island prospered under treaties that guaranteed its sponge fishing and boat building industries, limiting taxation. By the 19th century it was one of the wealthiest sea ports in the Mediterranean – at its peak, some 500 ships vied for moorings, and the population was supposedly in excess of 20,000, leaving improbable grandeur all around the harbour. Today’s tourists can still buy those sponges – though, much like the island’s fridge-magnets, it does pay not to heed the packaging for local authentiticy. Ottoman rule was followed in 1912 by Italian occupation, and that control was formalised after the First World War. By 1943, Symi had suffered three decades of imported pasta. Then, on 8 September, Italy changed sides.

Armistice turned the Dodecanese into a scramble for regional, strategic dominance. The airfields and large Italian garrison on Rhodes made it the greatest prize, and that island quickly fell under German control, but Symi was just as valuable.

It offered a stepping stone to Kos and Leros – and these were places Germany could not simply write off: it desperately wanted to secure them, before Britain could set up bases. Accordignly, after that armistice, Hitler’s orders were to hold what remained and retake what was lost: to keep neutral Turkey from tilting towards the Allies, and protect what functioned as a very useful supply route into the Balkans. Symi was a very small piece of a very large anxiety, well worth the fight.

WHO’S WHO

On the Allies’ side, it’s useful to be define who we’re talking about in 1943, otherwise, every caïque – every wooden-hulled boat – in the harbour looks as though it’s manned by the same men. A trawl online shows several confusions of operations, and books exploring this history are few and far between – I’ve listed them, below.

The Special Boat Squadron – the SBS – was British, formed partly from SAS and Small Scale Raiding Force personnel. The Greek Sacred Band, or Sacred Squadron, was a separate force under Colonel Christodoulos Tsigantes, Greek-manned and British-trained. The Levant Schooner Flotilla was principally a Royal Navy body, running armed caïques and local schooners to carry raiders, agents and supplies between the islands and the Turkish coast. By April 1944 those three maritime threads – LSF, SBS and Sacred Squadron – had been drawn together under one name, the Anglo-Hellenic Schooner Flotilla. It’s an organisational story, in other words, not four names for the same handful of boats. The fourth outfit, the Long Range Desert Group, also indulged in various bits of active island-hopping – we’ll get to them in Part II.

On 17 September, Major Ian ‘Jock’ Lapraik arrived on Symi with M Squadron, SBS. Twenty-six men, who promptly mounted a captured 20-mm Breda cannon on the ridge dividing Symi Town and Pedi village, and set themselves to better-organising the island’s defences in general. The Italian garrison numbered some 140 men, whose relationship with the Greek population was understandably strained after thirty-one years of occupation. Pasta isn’t that popular, here.

Lt. Commander Corrado Corradini had arrived from Rhodes to help pour olive oil on troubled waters, and rag-tag stragglers fleeing from Rhodes appeared from time to time, to swell the numbers further. But further reinforcements arrived by accident as much as design: intercepting a French sloop bound for newly invaded Kos, Lapraik diverted forty RAF ground crew of 74 Squadron onto Symi instead.

There was no doubt, Lapraik said, the island would shortly be attacked. It was only a question of when: his first Order of the Day to those men compared Symi’s position to Singapore, just before the Japanese arrived. Everyone had to be ‘on the job, all the time’, and men who’d trained to service aircraft became infantry, serving the SBS.

Patrols were the next thing. Every man set to learning Symi at ground level rather than merely seeing it on a map: taking routes over ridges to gauge distance through time-travelled, testing observation points, and assessing the many coves and beaches to work out which were landing places, and which were killing grounds.

That transition is easy to overlook. The SBS were raiders, navigating with charts that had been in circulation for decades to operate on ground that had only recently been drawn accurately, on a map.

In other words, men had to discover what contours meant in reality – but Germany was also deciphering the Dodecanese. Kos had fallen at the beginning of October, too quickly, too easily, and too close for comfort. The strategy in Berlin shifted towards offensive operations, with more of the islands earmarked for occupation, and – with admirable logic – by deciding it might be helpful to have radio stations on Naxos and Symi to support them. What German intelligence hadn’t quite grasped was that the SBS had beaten them to it.

PEDI, PEDANTIC

The Germans came for the island of Symi at dawn on 7 October. On the true size of that landing force, historians disagree with all the veracity of Doolittle’s double-headed llamas, all certain they’re heading in the right direction: German records put it at around 40 men in two caïques and a schooner; some British accounts have it as a 100 men; Thomas Harder has done an in-depth study of Anders Lassen – on Symi with Lapraik – and is confident, 59 Germans came ashore in Pedi bay in a caique, the Esperia, under one First Lieutenant Fresemann.

Pedi bay is an amphitheatre, wrapping its arms around a stutter of shuttered houses at ground level. Symi town lies to the west, its terraces rising steeply from the shore to the south of the harbour – up to a church and citadel. Any force landing at Pedi has entered its own, private theatre – overlooked from the stalls – and there sit the tragi-comedians that provide one of the classic ‘really?!’ moments in this little corner of the war. Gavin Mortimer tells it best. I’ll paraphrase:

When Andy Lassen woke at first light he looked down at Pedi bay, spotting uninvited guests. ‘Who the hell are they?’ he roared, to which the aircrew, standing guard, replied: ‘Well, they’ve just sort of landed’.

The RAF’s version of events was sanguine in admitting that Flight Sergeant Charlie Schofield, manning the Breda cannon, ‘wore spectacles on account of his short sight’. For his own part, Schofield insisted that the German landing party had been engaged the moment they hove into view – and yet somehow, 40 men managed to disembark and start moving inland towards Symi town, totally unshot-at.

As one SBS patrol moved to counter them, it came under fire from a German machine gun that had already found a vantage point above the houses; the Germans were quick to discover what anyone who visits Symi soon realises: a few metres of height dominates an astonishing amount of ground.

Back at Pedi, Schofield tried to keep the remaining Germans pinned down just off the beach: three Stukas had appeared, coming in low – which caused him to laugh, and reassure his comrades that their angle of approach was far too tight to be useful, and their hilltop position was safe. That said, by now the Breda cannon’s vibration had shattered his spectacles and the sights had fallen off the overheated weapon – short sighted or not though, he stuck to his guns, literally, and that courage was rewarded with a Military Medal.

Elsewhere, Lassen – already suffering a burn to his leg and dysentery – was on his way to a second bar for his Military Cross. He led an Italian counterattack that drove the remaining Germans back to their boats, and his citation claims 16 Germans killed, 35 wounded and seven captured, against one British dead and one wounded. That probably overstates the German toll. Some accounts put it nearer two dead, several badly wounded, and nine missing. What everyone agrees on though, is the shape of the day: a German withdrawal under fire, heads down as Lapraik sent an SBS caïque out to rake them, rounding the headland.

It was terrain, as much as anything, that scuppered their plans – the ridge obscuring Pedi village from Symi town had also hidden British positions from the sea, and men who’d spent weeks walking that ground had been in a position to use it as men who’d only glanced at a map could not. It was an impressive defensive success. But it was also temporary – the next day, more Stukas arrived. This time, not flying quite so low.

BOMBED OUT

General Ulrich Kleemann, commanding German forces in the Aegean, sent Stukas over Symi at nine the next morning, returning at two-hourly intervals for the rest of the day. Lance-Corporal Robert McKendrick of the SBS was killed; roughly two dozen Greek civilians died in the town; a direct hit on Lapraik’s headquarters brought down its makeshift hospital, trapping two SBS men, Tom Bishop and Sydney Greaves, beneath the rubble. In a short, grim episode that’s left a mark with everyone who’s written about that week, Ferris – suspended upside down into the wreckage, with little more than a scalpel and a pair of scissors – amputated Bishop’s trapped foot to free him. Both men died; Greaves under the debris, Bishop shortly after from shock and blast. For what they did that day, Ferris was awarded the Military Cross; the American medic, Porter Jarrell, received the George Medal.

The British could fight Germans in Pedi bay, and into Symi town, but they could little about Stukas strafing the island. A large number of houses were destroyed: their footings remain in the same, Luftwaffe-heavy condition now – several have recently attracted hopeful FOR SALE boards – and many civilians were killed. Lapraik finally withdrew what remained of his force on 12 October. Without adequate air cover, Britain’s position in the Dodecanese was increasingly precarious – and it wasn’t long before the Germans occupied Symi proper.

Nine months later, on the night of 13–14 July 1944, British and Greek special forces returned in strength. This was Operation TENEMENT: a force that crossed to Symi from the Turkish coast in launches, schooners and caïques, dividing into groups and coming ashore under darkness. 100 or so men of the SBS under Lapraik, with 224 men of Tsigantes’s Greek Sacred Squadron. By then Symi was no longer simply a shape on a map. British and Greek forces had spent months moving among the islands, turning contour-lines into recognisable waypoints. The men moved towards their objectives, the harbour defences and fortified ground above Symi town – climbing to positions from which they could dominate German and Italian posts below.

At dawn, mortars and machine guns opened fire; fighting spread around the harbour and through the terraced heights; by the time the force withdrew, twenty-one enemy troops had been reported killed and 151 prisoners taken, while vessels, weapons and installations had been destroyed. TENEMENT had no intention of holding Symi. Its aim was for men to overwhelm the garrison, destroy what they could, and disappear – which they did, with aplomb.

In 1943 and ‘44, SBS had relatively recent mapping of Symi to hand, and there’s no reason to diminish those map sheets’ quality to make the point that only took me only seconds to realise – ice cream purchased – stepping off the ferry last week.

The official wartime history of British military survey shows how uneven the region’s mapping inheritance had been to date. It’s true, GSGS had prepared some mapping of mainland Greece at 1:250,000 and 1:100,000 by gathering information from the Greek geographical service. Symi makes for an intriguing contrast. The excellent 1:25,000 map sheet was ultimately derived from Italian surveys – precisely the kind of large-scale mapping its military geographical service could deliver par excellence. It records the island’s interior accurately, though it cannot communicate the steepness under your legs, the dead ground your eyes find for themselves, the way heat rearranges distance, the way loose rock slows you down, and the precise way in which a single ridge can hide an entire village and harbour from a town, just a mile away.

A good military map had reduced a violently three-dimensional island into something that could be folded, copied and planned over, but the SBS and the Sacred Band still had to discover, on foot, what tidy contour lines actually meant. But there is another part to this story, one that takes us away from Pedi village and Symi town – and the knee-punishing Kali Strata; 385 polished, pine-resin-covered stone steps – over to the south-west of the island.

On the same 1:25,000 map I held last week, those painstakingly-drawn contours depict a magnificent, enclosed bay. At its head, sits the monastery of the Archangel Michael at Panormitis – a seat of prayer, crucial to the struggle for insight, observation and control of the Aegean. That is where Part II will begin.

In the meantime, this is a must-have if not exclusive list of bedtime reading:

On the SBS and the Aegean raiding forces

Gavin Mortimer, The SBS in World War II: An Illustrated History — an excellent tome, a great overview of the region’s campaigns, superbly illustrated, and useful for keeping the wider SBS chronology straight around the Symi episodes.

John Lodwick, The Filibusters: The Story of the Special Boat Service, reissued as Raiders from the Sea: The Story of the Special Boat Service in WWII — the first official history of the SBS, written by a man who served in the Aegean himself. Feels dated now, but it’s useful and timely; written in 1947.

Nicholas Jellicoe, George Jellicoe: SAS and SBS Commander — a biography of the man who actually ran Raiding Forces HQ and the Aegean operations that Lassen’s detachments answered to. Useful for the command-level view above Symi itself.

On Anders Lassen himself

Thomas Harder, Special Forces Hero: Anders Lassen VC MC — a rich biography of the man, well worth tracking down.

Mike Langley, Anders Lassen VC, MC, of the SAS — a biography drawn substantially from the testimony of the men who served alongside him.

On Symi, on the Sacred Squadron, and Levant Schooner Flotilla

Adrian Seligman, War in the Islands: Undercover Operations in the Aegean 1942–4, Recalled by Men of the Levant Schooner Flotilla, the Greek Sacred Company and Aegean Raiding Forces — possible the best-targeted book for this region. Seligman ran the Levant Schooner Flotilla that ferried raiders like Lassen’s men between islands including Symi, and this book is built from first-hand accounts.

W. E. Benyon-Tinker, Dust Upon the Sea (Hodder & Stoughton, 1947) — is the better single book for the flotilla. Benyon-Tinker was the flotilla’s intelligence and gunnery officer, and this a contemporary memoir of the operations, written just after the war,. It’s out of print — and no, I’m not lending you my copy.

Anastasios G. Vlachostathopoulos, Ieros Lochos 1942–1945 — is in Greek.

Academic bits and pieces

There’s a Naval Postgraduate School thesis, Amphibious and Special Operations in the Aegean Sea 1943–1945, that’s downloadable from various repositories. There’s also a shorter academic paper specifically titled The Sacred Squadron, on the Greek unit’s Aegean service, that’s on Academia.edu — which has become a tad unwieldy.

On the wider Dodecanese campaign

Anthony Rogers, Kos and Leros 1943: The German Conquest of the Dodecanese (Osprey Campaign series) — is focused on Kos and Leros rather than Symi, but it’s useful for the operational framework inside with the Symi raids sit. Keep your eyes open for Anthony’s forthcoming book via Osprey books in 2027, The Long Range Desert Group: Italy, Greece and the Balkans, 1944–1945.

I like this map of Valletta, which was first published in 1943 — or so the caption says.

But what draws my eye most isn’t the orderly grid inside the city walls — houses and streets dating back to the mid 1500s, laid out in obedient lines. No, it’s the rectangle: a small blank space on the waterfront below the Marsa, where the harbour narrows toward French Creek. It’s labelled ‘Floating Dock’.

Before I get to that though, the map sheet it’s sitting on deserves a word, because the word ‘map’ itself should raise a few eyebrows.

Technically, this is City Plan, MALTA 1:7,500, First Edition. Prepared, as was usual, by an un-named draughtsman working under the direction of the Chief of Engineers, U.S. Army, in 1943 – and copied from the British Admiralty’s Chart No. 974. One sheet of paper, three separate ways of getting dressed for the same event: the Admiralty’s hydrographers called their version a chart, because it was meant to keep a ship off the rocks; the U.S. Army Map Service called their tracing a plan, because it was meant to move a soldier round the city without getting shot; and I’m calling it a map, because that’s how most of us see pieces of paper like this one — eighty-five years after the fact, it seems churlish to call it anything else.

*Surveyors have opinions about this: a map orients you, a plan lets you build something, and a chart is a tool on which somebody’s safety depends, usually while you’re afloat.

Back to those two words on this map of Valletta, though.

A floating dock, unlike most things worth putting on a plan, doesn’t have one address. It’s a colossal steel box, built to be flooded down, walked out from under a ship, towed to wherever the work is, and pumped dry again wherever it lands. And depending on which creek had the cranes free or the depth a particular repair needed, Malta’s dock sat in more than one regular spot. So whoever surveyed it first for the Admiralty’s contribution to charting collateral didn’t just commit to a single rectangle — they added a note, in effect: if it isn’t here, try over there.

A small piece of honesty, but I like it. A lot. It’s an official aside, that the thing being mapped simply might not be where it was supposed to be. Except — that note can only be true of a dock that still moves, and by the time this particular sheet went to print, Malta’s didn’t.

Malta had no floating dock at all until 1925. Before that — for a naval base with what the Victorians liked to call ‘incomparable harbour facilities’ — there was nowhere to lift newer, larger warships clean out of the water for repair. On 17 July, the tugs finally towed the appropriate piece of infrastructure into Grand Harbour: Admiralty Floating Dock No. 8, a German-built piece of engineering handed to Britain as war reparations — one empire’s reparations bill turned into another empire’s dry dock. It did the job for fifteen years, presumably drifting between its usual berth and its alternative one as required, without anyone thinking to remark on it.

Then, on 20 June 1940, ten days into Italy’s war against Britain, Italian bombers hit it twenty-four times and it sank where it stood.

The floating dock didn’t drift anywhere after that. It stayed where it went down for twelve years, until the port authorities finally brought in an Italian floating crane to raise the wreck for scrap. Malta had no floating dock of its own for the rest of the war, and three years past it — which means that by 1943, ‘Alternative position’ had stopped being a useful note and, unintentionally, become a kind of fossil: an artefact from whichever earlier edition of Chart No. 974 had first recorded the dock’s two working berths, copied faithfully onto a wartime plan by people who had no reason to know it no longer applied.

The dock wasn’t at either position. It wasn’t floating at all. It was sitting, in a single, entirely un-alternative position, on the bottom of the harbour, and would go on sitting there for another nine years. It still makes me smile, though. A note meant to capture a moving object’s one flexibility — it might be here, or it might be there — outlived the thing it was describing, and was then reprinted, in good faith, years after both positions had become equally wrong.

For kicks, by the way, here’s GSGS 3859 showing Malta. To zoom in, pop over to Barry Rudderman’s site - smashing chap, awesome map collection.

The most peculiar thing about maps, is that once printed, they’re always out of date. In wartime, doubly so. A factory might be razed to the ground the day after ink hit paper; a track might have become a road due to heavy traffic; a bridge might no longer ford a river, and a dam might have been blown to smithereens. That map of Valletta was, technically, trying to be twice as careful as it needed to be, to be accurate for those about to use it. Still wrong, as it tried so hard to be right.

An added bonus: here’s a super example that shows how easy mapmakers of all kinds could get things wrong: an illustration for the Los Angeles Times in 1942, depicting the bombing of Malta. The eagle-eyed among you will note the mapmaker has Axis aircraft attacking from the southwest, which rather skews the footnotes of history.

Amazing, isn’t it, what you reveal in history when you study a bit of geography.

The most memorable comment on maps in the Burma campaign, has to be the wise words of Field-Marshal Sir William Slim. Writing in his 1959 book Unofficial History, he reflected that battles always seemed to happen “in the dark, uphill, in the rain, and at a spot that was inevitably at the corner of four separate map-sheets.”

It’s good, because it’s true. Anyone who’s tried to read a map in bad weather under fire will recognise it: someone has the wrong map, someone else has the right one but they’re at least fifty feet away, and the place that matters most is probably not on that map anyway, but the next map along. Yet while it’s a soldier’s observation, it’s also an odd thing for Slim — in particular — to have said.

Here was a man with a reputation as one of the most practical, unpretentious generals Britain had produced to date — famous for wanting to know how a road was built or a ration-lorry loaded, not just whether it turned up on time. But the only thing he said about maps, was disparaging. It’s a familiar habit of mind though, and not really his fault: people who read maps rarely ask how they were made. Even, it turns out, the people whose job it is to read them for a living.

Unofficial History is what it says on the tin — a set of after-dinner reflections, told at a comfortable distance from the paperwork, in the voice of a man entertaining a room rather than compiling a report. Very quickly, we get the idea that mapping in Burma was thin at best, and inconvenient at worst. Mention maps and the Far East today, and it won’t be long before someone volunteers more or less exactly that: the impression of a remote country, dense jungle — and presumably terrible maps. It has settled into the footnotes of history as something nobody bothers to check. Which is annoying.

Recently, I was asked for a specific map sheet: Survey of India 83 K/8, one inch to the mile, covering the country around Shangshak, east of Imphal. Sadly, I knew I didn’t have it, not a copy I could share — and more to the point, I didn’t have anything else from that K-block either, not a a decent scale. For reference, here’s Shangshak:

At this scale — one inch to two miles — it doesn’t inspire confidence. There’s a lot of dense green ink, it’s all pretty thick vegetation, and those contours are no joke.

Wartime maps for this theatre have been finding their way into institutional libraries and university collections for the better part of two decades. Searching them, and indeed, searching for them has never been easier: in minutes, a researcher can find material that would once have meant a train fare, a reading-room ticket, and a great deal of patience. What’s missing is a one-inch to the mile sheet for that ground — 83 K/8 — which is four times more detailed.

Same square of earth, much finer print: individual tracks, smaller features, the kind of resolution a battalion digging in on a specific hill would actually want, rather than the wider district view a half-inch sheet gives us.

So what happened there, and where is that mapping?

Shangshak is not obscure ground. It deserves to be better known. In mid-March 1944, the newly raised 50th Indian Parachute Brigade was sent into the hills north-east of Imphal to rest rather than fight. Nobody expected them to meet anything or anyone more serious en route than perhaps a long-range patrol. What they encountered was the leading edge of an entire Japanese division aimed at Kohima: a division that — much like the Spanish Inquisition, as I understand it — nobody had been expecting.

The brigade fell back on Shangshak — a Naga village that happened to command the tracks south-west to Imphal — and dug in as best it could. The ground didn’t allow for much digging: there, they found themselves on the rock of an extinct volcano, too hard for slim pickings, so what passed for trenches were barely scrapes in the topsoil. There was little potable water on the hill, either. Everything had to be carried up from the valley through a ring of Japanese lines that were tightening by the day, and more than one air-drop of supplies drifted down into enemy hands.

For the best part of a week they held on regardless, at a heavy cost, before those few who remained could break out and get to Imphal. Documents recovered later showed how much more of the Japanese force was heading for Kohima — and none of that is really my subject, and there are historians far more erudite than yours-truly who tell that particular story superbly. It’s the maps that interest me, the maps those men held in their hands.

The terrain, and what it took to survey it

The country itself is the place to start. The usual shorthand of jungle, mountains, and ridges is accurate without being remotely explanatory. Open a map of the Naga Hills instead. Ridge follows ridge, each one cutting off the view of the next; the valleys between them twist away almost as soon as they open; there are footpaths and tracks but few roads, and little that behaves like a straight line. A mile — an inch — on a piece of paper might mean a long descent through undergrowth with a bad attitude, a river crossing on a bamboo bridge that a good flood could remove overnight, and an equally punishing climb before the next ridge even comes into view.

But we do know how punishing it was, because this ground had been surveyed, and surveyed relatively well at a smaller scale, only a few years before the war reached it. Luckily for us, at least one of the men involved wrote a first-hand account.

In January 1937, Eric Lambert took the podium at the Royal Geographical Society in London to read a paper on the previous season’s topographic survey work.

In his own words, Lambert was ‘nominally an officer of the Indian Police’ — a humble understatement. He was a sub-divisional colonial officer and, at the time, a magistrate at Mokokchung when the Survey of India decided to send a triangulation party through his patch. The Government of Assam had judged it too dangerous to let surveyors loose in fully independent, tribal territory without a political officer who knew the ground. That meant a man in Lambert’s position had to travel with them, keeping the peace long enough for men with theodolites to do their work. One member of the audience couldn’t resist a good heckle: even in India, apparently, a policeman’s lot was not a happy one.

The whole point of that season’s survey work though, was to connect two existing chains of triangulation — one running up the Brahmaputra through Assam, the other up the Irrawaddy through Burma — by a third series, threading north-east across the Naga Hills, over the Patkai, and down toward the Chindwin. Maps did exist. It’s the scale and the detail of the map, that matters.

Major G. H. Osmaston led that survey party, with Lambert’s escort comprising four platoons of Assam Rifles under one Major L. C. Bull. The country beyond that district answered to no government at all, only to itself. The danger was not theoretical. Paths were seeded with panjis — sharpened bamboo stakes hidden under dead leaves, and almost in passing, Lambert noted that one of the villages was suspected of ‘the cutting up of a previous Survey Party’. That cutting up was literal.

The column moved in strict order — guide, scouts, advance guard, porters, baggage guard, porters again, rear guard — with temporary porters recruited only from the villages that lay ahead in the line of march. That way, there was much less opportunity for anyone to think better of the arrangement and vanish overnight.

Diplomacy was conducted the old-fashioned way: at one river crossing the local chiefs came down to meet the column and were given a bottle of rum and a hank of red wool in exchange for a mithun — a bit like a semi-domesticated small cow; less a working animal like a plough ox; more a wealth and status animal, sacrificed and eaten at feasts, and used as a formal gift to mark a meeting between groups. Owning mithun, and being in a position to give one away, was a marker of prestige. Most exchanges of mithun — including the ones between surveyors and chiefs — were marked by revelry and wild, wide-eyed dancing.

Two hundred Sema Naga porters carried the survey expedition’s stores, topped up as they went with temporary labour, because — as Lambert noted, with the particular deadpan of a man who’d had to solve this problem personally — a single porter would eat his own, entire sixty-pound load within a fortnight if nobody stopped him or resupplied him. Which rather defeated the purpose of sending anybody anywhere.

On one peak, the jungle on the summit was so thick, it took a hundred men three days simply to cut sightlines through it — three days of clearing, before a theodolite could be set up and aimed at anything. Multiply that across several hundred miles of range, in every direction, in a working season that’s bookended by monsoon cloud at one end and the smoke of hillsides burning at the other, and we can start to see why ‘surveyed’ and ‘well mapped’ were not, in this particular region, quite the same claim. None of which is to say the hazards were only physical. After Lambert presented his paper, it’s a safe bet the discussion that followed was a lively one.

During the session at RGS — a ‘this is more of a comment than a question…’ event — Colonel Rich, who’d spent three seasons in the early 1920s surveying the Hukawng valley and Chindwin watershed on the Burma side of the Patkai, noted that his own surveyors had returned with reports of human sacrifice and head-hunters. Rangoon dismissed those first reports as old wives’ tales; a second season’s investigation confirmed they were true, if rare. In one village, Rich’s men found a girl being kept for sacrifice that summer. Rich bought her freedom for £40, and sent her to missionaries at Myitkyina — where, in the detail I’ve found hardest to shake off, she was later given a hospital powder she mistook for poison, and drowned herself in the river rather than take it.

Lambert’s paper touched on the same ground: some of the villages being surveyed, he remarked, buried their dead beneath the household granary, and didn’t seem bothered if ‘buried’ was only ‘partly buried’. A former District Commissioner in the audience went further, noting that cemeteries could not be moved, and any growth in a village’s footprint simply meant that a street became a graveyard, so that walking through a settlement meant passing a run of houses, then a row of corpses old and fresh alike, and then houses again. Nobody was remotely troubled by the arrangement, and none of that detail will be found in any textbook on triangulation. Nor is it in the footnotes of military history. But it is, all the same, part of what a survey party in this country had to contend with on top of the jungle, the rain, the sixty-pound loads, and — once the fighting began — the risk of getting shot.

Lambert closed his presentation with this: “The Naga Hills — a beautiful country and a happy and lovable people. May they long be spared from the terrible consequences of Western Civilization.” Not, on the face of it, the sentiment you’d expect from a man who’d just spent a season helping the Government of India draw lines across those hills. But it’s a useful reminder, the people doing this work were rarely as simple as ‘surveyor’ or ‘policeman’ suggests. The maps resulting from those survey parties were never just academic abstractions — somebody had to dance for that mithun.

How the Army was supplied with maps

The organisation conducting those early surveys was not, in peacetime, a military affair. The Survey of India was a civil department, working in Circles — Shillong for Assam, independent parties for Burma and the south, a Frontier Circle up at Simla for the North-West. Britain had maintained a Geographical Section at the War Office since before the First World War to provide military mapping for its outposts; India had little comparable, no military survey representation at Army Headquarters — and this isn’t the ‘stack to offer a full history of the Survey of India.

Nor is it the place to set out the entire transition from civil affair to military shenanigins during the Second World War. Time, space, et cetera. Suffice it to say, when war reached India’s doorstep in 1942, the mapping units that would eventually supply the Fourteenth Army came into being — and were far from fully prepared.

The organisation was pulled together in a hurry: Survey of India staff transferred onto military establishments; field survey companies raised from scratch — one of which was reconstituted from a unit that had walked out of Burma with little but the clothes it stood up in, its triangulation records abandoned and lost at Myitkyina. Production units were shipped from England in lorries that had been built for European roads, to negotiate mountain passes that had never met a fourteen-ton vehicle before. Ink dried too fast. Paper faded and curled. Various, previously undocumented fungal contagions ate their way into the wooden legs of every piece of precious survey equipment — and that was the least of it.

Still, miracles happened.

Map depots were set up at Ranchi, Barrackpore and Bangalore — and the Barrackpore depot eventually moved into a converted jute mill with enough room to store five million map sheets, which gives some sense of scale.

Distribution ran by rail wherever the mapping could be guarded, and by liaison officer where it couldn’t — unescorted maps travelling into Assam had an unfortunate habit of simply not arriving. For issue to units in the field, closer to the frontlines, sheets were folded in three and bundled 50 at a time; 300 of those bundles filled a sandbag, 1,200 filled a flour sack, and a flour sack of maps weighed a solid 70 or 80 pounds — which is a detail I rather like. Somewhere in Assam in 1943, a fair number of men spent their war grumbling about backache. Maps — stacks of paper — are heavy.

One of the men who understood the whole supply chain, from theodolite to printing press, was Colonel — later Brigadier — Guy Bomford: seconded to the Survey of India in 1921, long before anyone was planning a fighting retreat from Burma, let alone a return to it. Perhaps, if you’d like it, I’ll do a series on people.

Working his way through a predictable career as a military surveyor, Bomford found himself occupied in postings across the Middle East, India, and Burma, and eventually responsible for the whole Eastern Army mapping effort. He later wrote what is, by agreement, one of the best standard English-language textbooks on the discipline of mapmaking — it’s fair to say he knew what he was talking about.

What he set down, in his own account of the Survey Service in situ, is precise rather than dramatic: by 1942, the whole of Burma and eastern and southern India had been surveyed at the 1-inch scale, ‘except for a belt covering the Chin and Naga hills where the ½-inch scale had been adopted’. It’s tempting to read that as a euphemism for ‘not done properly,’ but that’s unfair to the surveyors.

A half-inch sheet covers four times the ground of a one-inch one for the same cost in paper, engraving and — as Lambert’s porters would tell you — unbelievable human effort. In a country governed by ridgelines and passes rather than roads, a wider view arguably serves a commander better than a narrower one. What the coarser scale also tells us, reliably, is where the terrain itself made the finer scale prohibitively expensive. That belt runs directly over Shangshak.

From the summer of 1943, the drawing of any maps in that region fell to 63 (Indian) Reproduction Group, one of the units reinforcing the base production effort that was taking shape at Calcutta — their output for the year specifically includes new quarter-inch and half-inch editions of Arakan, Manipur and the Chin and Naga Hills.

By March 1944 — at about the same time the Japanese reached Shangshak — that effort had shifted. Survey directorate control of the Fourteenth Army had been passed to Eleventh Army Group, and it was now 71 (Indian) Base Map Reproduction Section at Harihar in southern India that formed the backbone of forward map production for the whole Burma campaign, along with air survey units working alongside it.

Even so, it was too little too late. While the 50th Indian Parachute Brigade was trying to dig in on a hill with very little water and, as far as I can establish, no good maps either, the unit that spent the rest of the war keeping Fourteenth Army supplied with maps was still getting itself together, several hundred miles to the rear at Harihar.

What’s missing now

Which brings me back to the missing sheet. The digitised Survey of India mapping I keep for reference runs to something like 9,500 — yes, nine and a half thousand — individual maps. One-inch and 50,000-scale sheets covering most of South Asia.

Grid block 83 — the block containing Assam, Manipur and the Naga Hills — is divided into sixteen quarter-blocks, lettered A through P. At the one-inch scale, A through J, L, M and N are all present and correct, several editions deep, running from 1909 into the 1960s. Burma’s neighbouring one-inch sheets, over the Upper Chindwin, are there too. But K — the block that holds Kohima, Shangshak and Ukhrul — is absent. So, for what it’s worth, is O, which sits next door. At the coarser half-inch and quarter-inch scales, both blocks survive: Kohima alone has quarter-inch editions from 1926, 1937, 1943 and 1945 — it’s only at the scale a battalion would actually have carried that this one patch of ground has gone quiet.

That’s no proof, the one-inch 83 K/8 doesn’t exist at that tighter scale. What’s more likely, is the basemap existed, was carried, folded, and refolded until its very substance gave way, and multiple copies didn’t survive the war — which was, for a working military map, the usual fate. Military mapping is paper that’s rarely left lying around — and the mapmakers’ stories, to date, have gone pretty much untold.*

None of that accounts for what’s sitting in private hands. I suspect that’s where the answer lies. Serious students of this campaign — and there are more of them, doing more careful work, than most people outside the field realise — tend to accumulate this kind of material over decades: given by veterans, bought at auction, inherited from a relative who served. Very little is catalogued, let alone digitised. I would be surprised if someone researching Kohima and Shangshak didn’t have an original Sheet 83 K/8, or something close to it, on a shelf somewhere. I keep a certain amount of paper maps myself, in addition to what’s indexed online, for exactly this reason.

Donations turn up, old bundles get re-catalogued, and shoeboxes get passed on. Sheet 83 K/8 may turn up tomorrow, in an archive that hasn’t finished cataloguing this decade’s donations. Or it may genuinely be gone — lost to eighty years of monsoon, mildew, and the happy accident of being merely just another piece of paper that needs filing somewhere. None of that changes the substance of Slim’s joke, which is still one of the best lines to come out of that campaign, and needs no correcting.

Would it have made a difference, having that one-inch sheet in hand?

Probably not. No map, however fine the scale, tells a brigade sent up to rest that a division is coming instead of a patrol, and no map puts water where the ground has decided there isn’t any. What the finer scale might have provided is harder to weigh: a trench line sited a little better on ground that offered almost nothing to work with, a resupply drop that lands inside the perimeter rather than drifting into Japanese hands, a track known in advance rather than found under fire. Shangshak was decided by surprise and by numbers, not by maps. But “wouldn’t have changed the outcome” and “wouldn’t have helped at all” are different claims, and I’m only confident of the first.

Long before four map sheets meet at an inconvenient corner, each one travels through survey parties in unmapped hills, through computing offices and engravers and printing presses, by rail as far as a railway goes and then by road or air or man or pack-mule — a map’s journey is rarely documented in detail. If it does its job well, then nobody doing the fighting ever has cause to think about it. Little wonder, few of us think about the maps, now.

If a stray 83 K/8 is sitting in your attic, I would love to hear about it. As to those untold mapmakers’ stories — well, you’re in the right place for news on that score.

Thanks for reading Mapping History.

Share

In all, Robert M. Chapin Jr. spent thirty-three years drawing informative maps for TIME Magazine — airbrushing dramatic mountains in grey-scale relief, sprinkling scarlet arrows, and stencilling tiny symbolic flags on top for good measure.

These maps weren’t made to confirm the front lines on any given Tuesday afternoon. Rather, Chapin’s maps were made to help a very frightened readership understand the much bigger picture — where the Second World War was going, and why.

Chapin came to mapmaking, sideways. In 1933, he gained a degree in architecture from the University of Pennsylvania — it was the height of the Depression, and quite possibly the worst moment in American history for building anything. That said, there was plenty of work for a man with a steady hand. Taking the first job he could, working as a photo retoucher for Newsweek, Chapin volunteered to draw a map of the River Nile — even though he’d never drawn a map before — and that, as they say, was that. From then on, he was the map chap.

Starting as a staff mapmaker*, TIME Magazine poached him in 1937, installed him in the Chrysler Building offices, and put him in charge of what would become one of the war’s most influential map departments. It started as a one-man operation, Chapin’s first assistant was his wife — Polly Sell Chapin, a fabric designer turned mapmaker — but by the time war had spread to all four corners of the compass, a team of six were producing the artwork for both TIME Maps and TIME Charts every week: researchers, artists, geographers, and draughtsmen.

*Chapin’s official title was Chief Cartographer. Strangely, the word ‘cartography’ wasn’t in common usage during the 1940s ~ it was foisted on the rest of the world by the French in the mid-1800s ~ so for most things WW2-related, I tend to use ‘mapmaker’ instead.

By December 1942 though, and with war now demanding as many as seven maps in a single issue, TIME’s publisher devoted a whole page to explaining who made its maps. It’s a wonderful piece of self-reporting, not least for how unglamorous it makes the whole operation sound:

“Right now this department is keeping five people very busy. Chief Cartographer is Robert M. Chapin Jr., who started his career as an architect and still has the architect’s gift of helping people to visualize a plan.

The tools were as distinctive as the results. An airbrush — ‘like a high-power atomizer’ — gave mountains, valleys and river beds three-dimensional weight. Two enormous globes, one political and one physical, were rigged from the ceiling on pulleys and counterweights, so they could be turned, lowered, and photographed from any angle, vital for the ‘global perspective’ that air and naval strategy suddenly demanded. And an entire library of celluloid stencils was created — bomb splashes, flags, Jeeps, and sinking ships — were created, so that no one had to redraw the same mini-explosions from scratch every week.

Behind it all sat a repository of some 1,500 maps, charts and photographs; a trove of topographic intelligence collected from various map libraries around the world.

U.S. Hydrographic Office charts and topographic base maps were found easily enough for most regions. Not always at a large scale, but then, Chapin’s work was mostly at a small scale, showing a vast area. In anticipation of the Allied landings, a range of Admiralty charts were acquired from work done by colonial British, French, Dutch, Spanish and Italian surveyors. But still, sometimtes, improvisation was the only way forward. Chapin’s map of Algeria, below, was drawn by combining insights from a French automobile roadmap with the hard-to-find ‘Atlas des Colonies Françaises’ — which says much about the poor state of those map libraries.

Chapin’s team also kept a map bank: basic maps drawn well in advance, ready to be finished at a few hours’ notice the moment the news caught up. This paid dividends: staring at a map of the Pacific and reasoning through where the next battle might take place, Chapin concluded it had to be the Solomon Islands. When U.S. Marines landed on Guadalcanal — TIME’s map of the battlezone was ready in hours.

The maps travelled far and wide. They were borrowed for reference works, reprinted at poster size for schools, lecturers, and Army and Navy officers around the world, and were — improbably — the first maps ever broadcast on television. In short, Chapin’s approach to explaining the bigger picture became a fixture in 20th century culture.

The team maintained its output for thirty-three years in total, making two or three maps most weeks, always on a punishing deadline. The process began at a Thursday story conference, where four or five candidate maps would be sketched out; two or three survived to be finished by 3:30 the next Sunday. His base maps were drawn on a projection table, letting him control scale and trace from the suspended globes for trickier perspective work. One of TIME’s articles described the process:

“The big problem for Cartographer R. M. Chapin Jr. and his staff was to get it all on a two-page map and still make it clearly readable... For the U.S. editions, the map was printed in eight colors — yellow, magenta, green, grey, gold, pink, blue and black. To get sharper differentiation between the lines and patterns, it was printed as if it were a piece of fine art — by sheet-fed offset on heavy paper — and then was bound with the rest of the magazine, which came off rotary letter presses.”

Colleagues described Chapin as ‘slim, relaxed, almost boyish,’ with an easy manner and a penchant for flying whenever he could — the best means to study the ground.

For his own part, he stated his ambition was to ‘dramatize the news of the week, not just produce a reference map like those in an atlas.’ Certainly, his legacy outlasted his time at TIME. Robert M. Chapin Jr. lay the groundwork for a fusion of storytelling that still shapes the way we consume media today — a bit like Substack.

A first substack is always a dilemma.
Stick around, and you’ll find me writing about all kinds of WW2 maps, and the men and women who made them. Illustrators like Chapin, yes, but more so the mapmakers who did more to help win this war than history has led us to believe.

If this is your kind of rabbit hole, then you’re in the right place.