The two most enigmatic structures of the western European Hercynian orogenic system are the (late Palaeozoic parts of the) Pyrenean mountain range and the extremely tight (in some estimates more than 180°) Cantabrian Orocline, the innermost section of the Ibero-Armorican Arc. They have developed coevally in close proximity to one another and any hypothesis proposed to explain the evolution of the western European Hercynides must explain both in terms of a single model. The Pyrenees have both a peculiar position in the Hercynian architecture and a strange evolution if viewed from the viewpoint of the development of orogens. They sit on the ‘hinterland’, south of the main orogenic body, the ‘innermost’ edge of which is marked by the south-vergent Montaigne Noire Helvetic-type nappes11The orogenic polarity used in this paper for the Hercynides is that they are, in toto and in contradiction to most existing models of a doubly-facing orogen, a solely north-facing (west and south-west in Iberia) orogen (see Robardet, 2002, 2003), although they have numerous and major elements, especially in their southern parts, that verge southwards.. They have a possible minor and local early shortening (D1) accompanied by Barrovian metamorphism followed by a major ESE–WNW-directed stretching that created a pervasive flat foliation and recumbent gneissic nappes of the first genre (D2) accompanied by a major Buchan-type metamorphism. This latter phase also displays porphyroblasts indicating dextral motion along the range. It was followed by later shortening nearly 90° to the stretching creating mostly upright structures with an E–W striking foliation and a following crenulation cleavage also with indication of continuing right-handed slip along the entire system (D3 and its subphases). The Visean to Bashkirian ‘flysches’ developed from north to south across the entire mountain range plus its bounding plains to the north and south and no flysch production accompanied any later deformation. All the major structures of the late Palaeozoic range face upwards. Such an architecture and sequence of events create a picture that is odd in terms of an orogenic development leading one to suspect the presence here of what John F. Dewey called ‘spoof orogeny’, but seems in accord with a keirogen that created, in addition to extension, a high geothermal gradient and associated Buchan-type HT/LP metamorphism. It was during the later episodes of the strike-slip in the Pyrenees that the Cantabrian Orocline formed. It seems clear that a westerly escaping fragment between the Pyrenean keirogen and the left-later shear systems of inner Iberia indented the formerly N–S Iberian segment of the Hercynides and produced the extreme curvature of the orogen similar to the tectonic evolution of the Appenninic/Alpine/Dinaride Orocline around the Apulian indenter (Argand's African promontory), the Carpathian/Balkan Orocline around the Moesian indenter, Hazara and the Assam Oroclines22The Hazara and the Assam oroclines are so tight that they have been classified also as syntaxes by Suess. But not all syntaxes are also oroclines: For example, the NE Greenland syntaxis joins two branches of early Palaeozoic orogens that formed along margins that meet at a right angle. The active Aleutian and the Kuril/Kamchatka arcs also meet at a syntaxis that is not an orocline. Oroclines are also different from simple deflections (Bucher, 1933, fig. 18; note here, however, that the fixist Bucher used at least four oroclines, namely those of Calabria, western Alps, Carpathians and the Balkans, and only one true deflection, the Alpine-Carpathian passage, to illustrate the concept of deflection in this figure; also in the same figure his concept of syntaxis is entirely wrong) that form around curved margins. Oroclines must show strain indicating bending synchronously with or postdating orogeny. around the Indian indenter (Argand's Indian promontory). Such a model is not in itself inconsistent with models involving lithospheric delamination, but I find it hard to see how the highest supracrustal sedimentary rocks in the core of the Cantabrian Orocline could have been preserved if a lithosphere-detaching thickening had taken place here. Contrary to the earlier models of lithospheric detachment by thickening, even much of Tibet still preserves its thick lithospheric root.