108 - The Geography of France’s Presidential Elections
The frequently fascinating and highly recommended Catholicgauze (“a blog on geography, geographic thought, and cool geography links”) presents an interesting map showing the results of the first round of France’s presidential elections on April 22, 2007.
About a dozen candidates participated in that first round, but the three main contenders turned out to be Nicolas Sarkozy (right-wing; 31,2%), Ségolène Royal (left-wing; 25,9%) and François Bayrou (centrist; 18,5%). Sarkozy and Royal, the two best-scoring candidates, face each other in the second round on May 6, 2007.
The map shows which of the three main contenders came out on top in each of France’s 100 départements (91 in France proper, France’s 9 overseas départements are represented by dots on the left-hand side of the map). And in doing so tells a more interesting story than the mere percentage points mentioned above.
For starters, it shows how localised support really was for Bayrou, who only succeeded in winning his home département of Pyrennées-Atlantiques.
Sarkozy, on the other hand, must regret France has a proportional electoral system for the presidency and not a ‘first past the post’ one as in the US (where the biggest vote-getter in a state receives all the political capital - i.e. Presidential Electors - for that state). For he is the biggest vote-getter in 74 out of the 100 départements.
‘Sarko’ wins 6 out of 9 overseas départements (all 4 Pacific ones - New Caledonia, French Polynesia, Wallis and Futuna, Mayotte; and 2 in the Americas - French Guyana, Guadeloupe) and 68 out of 91 of the ‘metropolitan’ départements.
Royal wins in 24 départements - 3 overseas (Martinique and St Pierre and Miquelon in the Americas, and Réunion in the Indian Ocean) and 21 in France itself. Those are, aptly, left of centre (cartographically speaking).
The geographic distribution of Sarko and Ségo’s respective electoral strongholds is quite striking. Sarkozy holds sway unopposed north, east and south-east of the country. Royal dominates Brittany (the ‘nose’ of France) and the south-west, and won in two electoral exclaves in Sarko-territory: a part of Paris, and the département of Nièvre.
These exclaves notwithstanding, the areas in which each of the two main contenders won, are mainly contiguous, as if they were ’sub-countries’ gearing up for a confrontation. I am reminded of election maps of Ukraine’s most recent presidential elections, in which the west of the country voted for the Viktor Yushtchenko (who wanted to move his country ‘westward’, into the EU and NATO) and the east voted for Viktor Yanukovich (who was more oriented towards Russia, Ukraine’s eastern neighbour).
Catholicgauze mentions that the voting pattern this time around was similar to previous elections: “The east and north parts of France vote conservative while the southwest, the west, and Paris go Socialist.”
It would be interesting to find an explanation for this geo-electoral phenomenon. Mesdames et messieurs, vos commentaires, s’il vous plaît!
PS - I’m hopeless with numbers, so please excuse any discrepancy, inconsistency or fallacy in my tallying up of the départements. Especially since one département is left blank on this map, as I’ve just now noticed: Haute-Savoie, bordering Switzerland (just below Lake Geneva, to be exact).
PPS - Another oversight is the white dot towards the south coast of France: an exclave of the département du Vaucluse, just to the south of it.
just a bit of explanation for Nièvre département : it was François Mitterrand’s département (former left-wing President of the French Republic), and that’s the main reason why Ségolène Royal has won there..
as for the south-east of the country, there’s been an old tradition of “radical-socialism” (in French politics, it means left-center wing, not far-left) since the beginning of the republican regime at the end of the 19th century… and it still goes on today…
what is very interesting, but it’s been on its way since the 1980’s, is that the industrial part of the country, where the working class used to be important, has voted for the right wing… and also, for Le Pen (far right wing), but less than before…
and some of the agricultural parts have voted for the left wing…
it’s a sort of reversing of the traditionnal votes…
Comment by Benoit — April 25, 2007 @
In fact it’s not south east but south west.
The département of Haute-Savoie has chosen Sarkozy.
You can have the results on the official webiste of the government:
http://www.interieur.gouv.fr/sections/a_votre_service/resultats-elections/PR2007/index.html
And there’s one thong interesting. Look at the results of people who live out of France and you’ll see that their results are very different from the ones in the “metropole”.
Comment by JMA — April 25, 2007 @
Politically interesting to be sure, but I’m not sure this counts as a “strange map”.
Comment by Rubrick — April 25, 2007 @
Yeah sorry about blanks. I and the other cartographers on Wikipedia didn’t have all the numbers in at the time.
As for the South I have always been told there is still animosity towards the Parisians because of their cultural dominance. This has been reflected via movements as early as the Cathars up until Vichy France’s apathy towards Paris under German occupation yet Vichy “official” control.
Comment by Catholicgauze — April 25, 2007 @
the red departement on the NE of paris is in the former “red belt”, the suburb who was communist.
Comment by m. — April 25, 2007 @
Inside the ‘départements’ are ‘circonscriptions’, and this makes for a more precise map where you can see the city/country divide, and how paris is cut east/west.
http://www.lemonde.fr/web/panorama/0,11-0@2-823448,32-901390,0.html
Comment by Sophie — April 25, 2007 @
Actually, Royal came ahead in Brittany and some of the west because it’s a part where the right wing is strong bu “moderate”, and as such the right wingers split more evenly between Sarkozy and Bayrou, letting Royal ahead…
French voting patterns are much less divided on a national scale than in the US…
Comment by Linca — April 25, 2007 @
I disagree with a few of your statments, which all revolve around the fact that the map only shows who got the most votes, giving no information about how many votes anyone received.
“it shows how localised support really was for Bayrou”
No, it doesn’t. Bayrou could have received 15% of the vote in all 99 other départements.
“Sarkozy holds sway unopposed north, east and south-east of the country.”
The word “unopposed” is unwarrented here. Sarkozy may have received only 45% of the vote in some of these areas (image that Bayrou had 20% and Royal 35%). Indeed, in some of these areas, the majority of the people may favor someone other than Sarkozy. Infact, nearly one quarter of the people favor none of those three candidates.
“Sarkozy, on the other hand, must regret France has a proportional electoral system for the presidency and not a ‘first past the post’ one as in the US (where the biggest vote-getter in a state receives all the political capital . . .”
But even in the US you can become president winning only 11 states. Without population information, for all we know Sarkozy is winning the French equivilent of Wyoming and Montana.
While not really “strange”, the map is interesting. Keep up the good work.
Comment by Clark — April 25, 2007 @
One strange fact: most of the areas in the West and Southwest that went for Royal were strongholds of the Girondins and, well, Royalists back in 1793.
So it fits your blog to a T, I say…
Comment by Chris — April 25, 2007 @
@ Clark:
You are right, of course: dividing France into homogenous red and blue areas based on who won the most votes in each département is quite a crude method. Some nuance is provided by the further subdivision of départements into conscriptions (as can be seen via the link provided by Sophie in comment #6).
Similarly, the stark division of electoral America into the ‘blue’ coasts and the ‘red’ interior gets more muddled once you look at the county instead of the state level.
Yet the point of a ‘crude’ map like this one is that it does say something about national politics in France. In my opinion anyway, as the resulting areas seem a bit too contiguous to be entirely random. Maybe they only say something if one stretches the facts a bit - by leaving out that the results of other candidates could have been instrumental in determining who of the two main candidates got the most votes in a particular département (as pointed out by Linca in comment #7).
It will be interesting to see how the results stack up in the second round, and if the red-blue divide will change significantly, when there will be only two candidates to choose between.
Comment by strangemaps — April 25, 2007 @
It’s intriguing to see the geographic spread of votes, and the regional preferences of support.
That split doesn’t seem to be at all along the same lines as in the UK and US where it is the cities and industrial areas which vote for to the centre left whilst the countryside votes for the centre right.
For example, the ports of Provence and the former coalfields of the north and east both returned Sarkozy. In comparison, it’s hard to imagine places like Sunderland, Newcastle and South Yorkshire ever voting Tory.
Perhaps the support for the extreme right in the south of France might have an effect there, and indeed it would be interesting to see a map showing the distribution of LePen’s support.
Meanwhile, it’s notable that even in so-called ’socialist’ Paris, only the NE suburbs actually showed a majority for Royal. The situation in London is more or less reversed, with only a small number of the most affluent constituencies voting Conservative, at least in recent years.
It’s also quite instructive to read the comment that the third candidate’s support was much more widely spread than the map suggests, since this illustrates exactly the kind of problems faced by third parties in both the UK and the US where first-past-the-post electoral systems still operate.
Comment by Roads — April 26, 2007 @
I think the comments here show one of the problems with maps of this sort; if you attempt to show the data down to a very low level, it overwhelms you with information, while if you look at a higher level you can see trends, but lose some of the subtleties. Sometimes the trends are what you’re looking at, and sometimes it’s the details.
The combination of the one above, and the one that Sophie linked to show us some interesting trends. If you try and look at all the individual maps of each candidate on the linked map, there’s so much information that you need to ask what you’re looking for to compare them meaningfully.
Comment by Neil W — April 26, 2007 @
Sarkozy… must regret France has a proportional electoral system for the presidency.
You can’t have proportional representation when electing one!
And a two-round majority system is not proportional. The eventual winner will get 100% of the presidency despite 30% or 27% of the initial votes (and maybe barely over half the second-round votes).
The map is interesting, but would it be possible to use shades of red/blue/green to indicate how strong the plurality was in any given department? In a multi-candidate race, just knowing who led doesn’t really show that much. And, as others have noted here, it doesn’t show the population disparities. As usual, the conservative does better the less dense the population.
I just discovered this blog (via a link at PoliBlog) and certainly will come back often.
Comment by MSS — April 26, 2007 @
A three-way race is well suited to RGB display: let each of the color channels represent the number of votes cast for one of the candidates in the territory represented by a pixel. This would show population density and the size of the minority vote.
Comment by Anton Sherwood — April 27, 2007 @
If you like maps of this sort take a look at the maps section of http://www.alba.org.uk/maps (the Scottish Politics site). For instance compare the Scottish results of successive UK general elections over the past 30 years or so at http://www.alba.org.uk/maps/generalmaps.html and you can easily see the quite awesome decline of the Conservative Party in Scotland (the disappearing blue one).
Comment by Chris Cooke — April 27, 2007 @
St Pierre et Miquelon? What’s that all about then?
Go Sego!
As for the Basques, they, by right, will always go for the different. All strength to them.
Keep up the good work
Comment by lordhutton — April 27, 2007 @
It would be interesting to find an explanation for this geo-electoral phenomenon.
I happen to have run across a possible explanation for this in my reading recently. This is from Diarmaid MacCulloch’s The Reformation: A History, pp 473-474.
The end of toleration in 1685 left a legacy of bitterness and instability in France, for it failed to destroy the Huguenots, while encouraging an arrogance and exclusiveness within the established Catholic Church. In the great French Revolution after 1789 this divide was one of the forces encouraging the extraordinary degree of revulsion against Catholic Church institutions, clergy, and religious that produced the atrocities of the 1790s; beyond that it created the anticlericalism which has so characteristic of the left of modern southern Europe. In the history of modern France, it is striking how the areas in the south that after 1572 formed the Protestant heartlands continued to form the backbone of anticlerical, antimonarchical voters for successive Republics , and even in the late twentieth century they were still delivering a reliable vote for French Socialism.
Comment by J. Dunn — April 28, 2007 @
Please may be intersting if I live here the text of the intervent of Gerald Batten to European Parliament concerning “Prodi our Men”?
Seems that is waited a video partecipation of Prodi to a Madame Royale Convention.
It seems to me not to be a beautifull things.
G.Schiavi from Roma
GERARD BATTEN MEP
‘MATTERS OF POLITICAL IMPORTANCE’
60 SECOND SPEECH TO EUROPEAN PARLIAMENT
25th April 2007
Joint Debate - Human Rights in the World in 2006 and the EU’s policy on the matter.
On the subject of Human Rights I would like to draw attention to the plight of a political prisoner in the European Union.
He has been imprisoned in Rome for the last four months.
He is now in poor health both physically and mentally.
He is being held without prospect of release or trial to try and break his will in an attempt to force him to sign false confessions against himself and others.
His name is Mario Scaramella, and his alleged offences are contrived accusations without foundation.
Mr Scaramella was of course the man who went to London in November 2006 to warn Alexander Litvinenko that he was about to be murdered.
Mr Scaramella and Mr Litvinenko had both been associated with the Mitrokhin Commission investigating links between Italian politicians and the KGB.
Mr Scaramella should be released and returned to his family immediately pending any trial.
Comment by G.Schiavi — April 28, 2007 @
A couple of rectifications:
96 departements in the ‘Metropole’
(Corsica is divised into 2 departements) but only 4 overseas departements Martinique, Guadeloupe, Guyane, Reunion
The rest are overseas territories with a different administrative framework.
https://www.cia.gov/cia/publications/factbook/geos/fr.html
CU
X
Comment by xavier — April 30, 2007 @
This is irrelevant to this particular post, but i can’t find any other way to contact you, so:
http://xkcd.com/c256.html
Comment by Tom — May 2, 2007 @
Sorry for OFF TOPIC:
This will certainly interest you. A map of online communitys:
http://www.notcot.com/images/online_communities.png
By the way, I really like your blog as I like (strange) maps!
Comment by Fireball — May 2, 2007 @
Here is the results of the second turn.
http://www.lemonde.fr/web/infog/0,47-0@2-823448,54-904808,0.html
Comment by JMA — May 7, 2007 @
this is another interesting map: operation radius of a WWII “New York bomber” Me 264http://letectvi.wz.cz/nemecko/me264/me264_7.jpg
Comment by JAn — May 7, 2007 @
“it shows how localised support really was for Bayrou”
No, it doesn’t. Bayrou could have received 15% of the vote in all 99 other départements.
As this poster already pointed out, the fact that Bayrou won only one departament shows exactly the opposite - how geographically spread it was. I’m stumped on how you would make the opposite conclusion from this simple fact.
Comment by Krum — May 7, 2007 @
This is a very interesting discussion, and I want to thank the ur-blogger for calling attention to the split. I think that the suggestion of J. Dunn that the divide is in part a reflection of the historic animosity of the Hugeunots towards the established church that has carried over to the political establishment is very intriguing. I would like to amplify that comment by mentioning that the southwest of France has a history of separation from the rest of the country that is much older than the Hugeunots. Specifically, in Roman Gaul, this area was a separate province, called Aquitania. After the fall of the Roman Empire in the West, it was the only area of modern France that was occupied for any substantial period of time by The Visigoths, who also occupied Spain. Perhaps most importantly, after the marriage of Eleanor of Aquitane to Henry II of England in 1152, it was a possession of the English crown to a great a lesser extent for three hundred years, until 1453.
So what, one might ask? Well, one difference this made was that a centralized government was not established in this area of France until much later than in most of the rest of France, since the English monarchy was never as able to do so as the French monarchy.
The relative independence of Brittany, the other region where Sego won a plurality, goes back just as far. In fact, the name of the area in French, “Bretagne”, is a reflection of all the Celtic refugees from Britain (Grandes Bretagne in French) who fled the Germanic invasions of what is today (but was not then) called England, starting around 450. The modern Breton language is a Celtic language, not a Latinate language like French or Provencal. The current Breton regional anthem is set to the same tune as the Welsh and Cornish anthems.
Like the Aquitaine, Brittany was the site of warring between the French and English crowns during most of the Middle Ages, but it was rather more successful in playing the two off against each other and thereby largely maintaining its independence until 1532. It maintained its own Parlement until 1789 and was a center of royalist resistance to the French Revolution.
The great difference between the two areas is that Brittany was and remains one of the most resolutely Catholic areas of France. So maybe the current voting pattern has more to do with general alienation from traditional French authority than with specifically an anticlerical variety of alienation.
There are similar (very) long term geographical patterns that may be behind voting in the UK and Spain (such as the historical radicalism of Andalusia).
Just my two cents worth.
J. Everett
Comment by J. Everett — May 7, 2007 @
good write thanks
Comment by sohbet — May 7, 2007 @
To the first comment about the red exclave of the nièvre and Mitterand:
I would interpret it the other way around. Mitterand was elected there because he had chosen to candidate in a stronghold of the left. It helped him to stay elected through the up and down of his career.
Why the nievre is a stronghold of the left, I don´t know. Maybe an early industrialization (mining) as a potential explanation?
Comment by info — May 15, 2007 @
[...] excellent map of the new political [...]
Pingback by THE FINAL MAP OF THE LAST FRENCH PRESIDENTIAL ELECTIONS at WHAT’S NEXT: INNOVATIONS IN NEWSPAPERS — May 25, 2007 @
Did you see that tiny red speck amidst the sea of blue. Now that the town I want to visit to see what makes it so different.
Comment by Phoenix Arizona Auto Home Life Insurance — June 29, 2007 @
That red speck is the Seine-Saint-Denis, a gritty area northeast of Paris inhabited largely by impoverished immigrants. Not so sure you really want to be visiting it.
Comment by jm — July 10, 2007 @
Umm, the US doe snot have “first past the post” Individual states may regulate how their electoral college reps vote. Indeed, several states require that the states votes be divvied as closely as possible to the proportions of the popular vote.
Comment by belg4mit — July 30, 2007 @
It fits chernobyl’s radioactive effects on the population.
http://www.futura-sciences.com/uploads/tx_oxcsfutura/img/retombee_tchernobyl_carte.jpg
Comment by koyotl — September 28, 2007 @
[...] or towns that dress up as French or have French heritage, this place is in fact part of France (see how they voted in the recent French election!) just 800 miles Northeast of Boston. I must [...]
Pingback by limeduck » “In order not to go blind, you have to travel” — November 17, 2007 @
[...] The rest of the country is two-kisses territory, apart from the same département in northeast Paris that stood out by turning Royal red amidst a sea of Sarkozy blue in the first round of the French presidential elections earlier this year (see entry #108). [...]
Pingback by 210 - French Kissing Map « strange maps — December 2, 2007 @
[...] The rest of the country is two-kisses territory, apart from the same département in northeast Paris that stood out by turning Royal red amidst a sea of Sarkozy blue in the first round of the French presidential elections earlier this year (see entry #108). [...]
Pingback by French Kissing Map « Later On — December 2, 2007 @
more results for Brittany
http://www.udb-bzh.net/article.php3?id_article=450&lang=fr
Comment by Gwenael HENRY — July 19, 2008 @