Contents of this chapter
1. Nonnulli antiqui architecti negaverunt dorico genere aedes sacras oportere fieri, quod mendosae et disconvenientes in his symmetriae conficiebantur. Itaque negavit Arcesius, item Pythius, non minus Hermogenes. Nam is cum paratam habuisset marmoris copiam in doricae aedis perfectionem, commutavit ex eadem copia et eam ionicam Libero Patri fecit. Sed tamen non quod invenusta est species aut genus aut formae dignitas, sed quod inpedita est distributio et incommoda in opere triglyphorum et lacunariorum distributione.
1. Some ancient architects have said that temples should not be constructed in the Doric style, because faulty and unsuitable correspondence arose in them; for example Arcesius, Pythius, and especially Hermomgenes. For the last named after preparing a supply of marble for a temple in the Doric style, changed over, using the same marble, and built an Ionic temple to Father Bacchus not because the form or style or dignity of the plan is displeasing, but because the distribution of the triglyphs and soffits is confused and inconvenient.
2. Namque necesse est triglyphos constitui contra medios tetrantes columnarum, metopasque, quae inter triglyphos fient, aeque longas esse quam altas. Contraque in angulares columnas triglyphi in extremis partibus constituuntur et non contra medios tetrantes. Ita metopae quae quae proximae ad angulares triglypos fiunt, non exeunt quadratae sed oblongiores triglyphi dimidia latitudine. At qui metopas aequales volunt facere, intercolumnia extrema contrahunt triglyphi dimidia latitudine. Hoc autem, sive in metoparum longitudinibus sive intercolumniorum contractionibus efficietur, est mendosum. Quapropter antiqui vitare visi sunt in aedibus sacris doricae symmetriae rationem.
2. For if it is necessary that the triglyphs should be placed over the middle quadrants of the columns, and that the metopes which are constructed between the triglyphs should be as broad as they are high. On the other hand, the triglyphs aganst the corner columns are placed at their furthest edge, and not against the middle of the columns. Thus the metopes which are next to the corner triglyphs do not come out square but oblong by half the breath of a triglyph. But those who wish to make metopes equal contract the extreme intercolumniations by half the breadth of a triglyph. Whether the work proceeds by lengthening the metopes or contracting the intercolumniations, it is faulty. Hence the ancients, as it seems, avoided the Doric order in temples.
3. Nos autem exponimus, uti ordo postulat, quemadmodum a praeceptoribus accepimus, uti, si qui voluerit his rationibus adtendens ita ingredi, habeat proportiones explicatas, quibus emendatas et sine vitiis efficere pssit aedium sacrarum dorico more perfectiones. Frons aedis doricae in loco, quo columnae constituuntur, dividatur, si tetrastylo erit, in partes XXVII, si hexastylos, XXXXII. Ex his pars una erit modulus, qui Graece embater dicitur, cuius moduli constitutione ratiocinationibus efficiuntur omnis operis distributiones.
3. Now we, following the arrangement demanded in accordance with the instruction of our masters, proceed in such a way that, if the reader will conform to our methofs, he may find those proportions set forth by which he can carry out temples in the doric style faultless and without blemish. The front of a Doric temples is to be divided along the line whre columns are set, into 27 parts if it is tetrastyle, into 42 parts if it is hexastyle. Of these one part will be the module (which in Greek is called embater) and when this is determined, the distribution of all the work is produced by multiples of it.
4. Crassitudo columnarum erit duorum modulorum, altitudo cum capitulo XIIII. Capituli crassitudo unius moduli, latitudo duorum et moduli sextae partis. Crassitudo capituli dividatur in partes tres, e quibus una plinthus cum cymatio fiat, altera echinus cum anulis, tertia hypotrachelion. Contrahatur columna ita, uti in tertio libro de ionicis est scriptum. Epistylii altitudo unius moduli cum taenia et guttis; taenia moduli septima; guttarum longitudo sub taenia contra triglyphos alta cum regula parte sexta moduli praependeat. Item epistylii latitudo ima respondeat hypotrachelio summae columnae. Supra epistylium conlocandi sunt triglyphi cum suis metopis, alti unius (et) dimidiati moduli, lati in fronte unius moduli, ita divisi, ut in angularibus columnis et in mediis contra tetrantes medios sint conlocati, et intercolumniis reliquis bini, in mediis pronao et postico terni. Ita relaxatis mediis intervallis sine inpeditionibus aditus accedentibus erit ad deorum simulacra.
4. The diameter of the columns will be two modules, the height including the capital 14, the height of the capital is one module, the width two modules and a sixth. The height of the capital is to be divided into three parts, of which one is to be the abacus with the cymatium; the second the echinus with fillets; the third the necking. The column is to be diminished as directed for the Ionic order in the third book. The height of the architrave is to be one module including the taenia and guttae; the taenia is to be the seventh part of a module; the length of the guttae under the taenia corresponds to the triglyphs, and is to hang down, including the fillet, the sixth part of a module. The breadth also of the architrave at the soffit is to correspond to the necking of the column at the top. Above the architrave are to be placed the triglyphs with metopes; the trigplyphs being a module and a half high and one module wide in front, and so distributed that in the columns both at the corners and in the middle, they are placed over the centres. In the middle, they are placed over the centres. In the middle intercolumniations of the front and back there are to be three, in the other intercolumniations there are to be two triglyphs. The middle intercolumniations are to be thus widened so that for those who are approaching the statues of the gods there may be an uninterrupted approach.
5. Triglyphorum latitudo dividatur in partes sex, ex quibus quinque partibus in medio, duae dimidiae dextra ac sinistra designentur regula. Una in medio deformetur femur, quod Graece meros dicitur; secundum eam canaliculi ad normae cacumen inprimantur; ex ordine eorum dextra ac sinistra altera femina constituantur; in extremis partibus semicanaliculi intervertantur. Triglyphis ita conlocatis, metopae quae sunt inter triglyphos, aeque altae sint quam longaea; item in extremis angulis semimetopia sint inpressa dimidia moduli latitudine. Ita enim erit, ut omnia vitia et metoparum et intercolumniorum et lacunariorum, quod aequales divisiones factae erunt, emendentur.
5. The width of the triglyphs is to be divided into six parts, of which five parts are to be in the middle, and two halves right and left are to be marked by the length of the fillet. The part in the middle is to be shaped flat as the thigh (which in Greek is called meros). Parallel channels are to be sunk with sides meeting in a right angle. To the right and left of them, in order, other flat surfaces or thighs are to be put. After placing the triglyphs, themetopes which separate them, are to be made as high as they are long. Further, at the extreme corners, half metopes are to be made with half a module wide. Hence the divisions will be made uniform and all the faults, both of metopes and intercolumniations and soffits, will be removed.
6. Triglyphi capitula sexta parte moduli sunt faciunda. Supra triglyphorum capitula corona est conlocanda in proiectura dimidiae et sextae partis habens cymatium doricum in imo, alterum in summo. Item cum cymatis corona crassa ex dimidia moduli. Dividendae autem sunt in corona ima ad perpendiculum triglyphorum et medias metopas viarum derectiones et guttarum distributiones, ita uti guttae sex in longitudinem, tres in latitudinem pateant. Reliqua spatia, quod latiores sint metopae quam triglyphi, pura relinquantur aut numina scalpantur, ad ipsumsue mentum coronae oncidatur linea quae scotia dicitur. Reliqua omnia, tympana, simae, coronae, quemadmodum supra scriptum est in ionicis, ita perficiantur.
6. The capitals of the triglyphs are to be made of the sixth part of a module. Above the capitals of the triglyphs, the cornice is to be placed projecting two thirds of a module, with a Doric cymatium below and another at the top. Further, with the cymatia, the cornice will be half a module high. Now in the lowest part of the cornice, above the triglyphs and the middle of the metopes, the lines of the viae and the rows of the guttae are to be divided so that there are six guttae in the length and three in the breadth. The remaining spaces, because the metopes are broader than triglyphs, are to be left plain, or divine images are to be carved; and at the very adge of the cornice a line is to be cut in which is called the scotia. All the rest - namely the field of the pediment, the cymas, the cornices - are to be finished as prescribed above for Ionic buildings.
7. Haec ratio in operibus diastylis erit constituta. Si vero systylon et monotriglypon opus erit faciundum, frons aedis, si tetrastylo erit, dividatur in partes XVIIII S, si hexastylos erit, dividatur in partes XXVIIII S. Ex his pars una erit modulus, ad quem, uti supra scriptum est, dividantur.
7. Such is the method that will be appointed for diastyle works. But if the work is to be systyle and with single triglyphs, the front of a tetrastyle temple is to be divided into nineteen and a half parts; of a hexastyle temple into twenty nine and a half parts. Of these one part will be the module according to which they are to be divided, as written above.
8. Ita supra singula epistylia et metopae et triglyphi bini erunt conlocandi; in angularibus hoc amplius, quantum dimidiatum est spatium hemitriglyphi, id accedit. In mediano contra fastigium trium triglyphorum et trium metoparum spatium distabit, quod latius medium intercolumnium accedentibus ad aedem habeat laxamentum et adversus simulacra deorum aspectus dignitatem.
8. Thus above the single architrave, two metopes and two triglyphs are to be placed. At the angles in addition, as much as is half a triglyph is put. Against the middle of the pediment, a space will intervene of three triglyphs and three metopes, in order that the middle intercolumniation, being broader, may give room to persons approaching the temple, and furnish a dignified appearance as one goes to meet the Image of the God.
9. Columnas autem striari XX striis oportet. Quae si planae erunt, angulos habeant XX designatos. Sin autem excavabuntur, sic est forma facienda, ita uti quam magnum est intervallum striae, tam magnis striaturae paribus lateribus quadratum describatur; in medio autem quadrato circini centrum conlocetur et agatur linea rotundationis, quae quadrationis angulos tangat, et quantum erit curvaturae inter rotundationem et quadratam descriptionem, tantum ad formam excaventur. Ita dorica columna sui generis striaturae habebit perfectionem.
9. The columns ought to be fluted with 20 flutes. If the flutes are flat, the columns must have 20 vertical edges marked. But if the flutes are hollow, we must fix their form in this way: draw a square with equal sides as great as is the width of the fluting. Now in the middle of the square the centre of a circle is to be placed, and let a circle be described which touches the angles of the square; and the curve which comes between the circumference and the side of the square, will give the hollow of the flutes. Thus the Doric column will have the fluting proper to its order.
10. De adiectione eius, qua media adaugatur, uti in tertio volumine de ionicis est perscripta, ita et in his transferatur.
Quoniam exterior species symmetriarum et corinthiorum et doricorum et ionicorum est perscripta, necesse est etiam interiores cellarum pronaique distributiones explicare.
10. Concerning the entasis of the column by which it is increased in the middle, the method prescribed in the third book for the Ionic order is to be imitated in the case of the doric order.
Inasmuch as the external appearance of the symmetries of the Corinthian and Doric and Ionic orders has been described, we must proceed to explain the interior distribution of the apartments of the temple, and also of the approach to the temple.
The Doric order is the most typical building order of the Greek mainland. It has direct references to the original wooden construction of the archaic temples. Claude Perrault, in his French translation and commentary of 1684, points already to this fact when he explains the origin of the guttae.
We may accept that the in the oldest temples the columns were of wood. To form the transition between the column and the entablature a simple cushionlike element was inserted. This became the Doric capital.
The entablature of the early temples has almost entirely disappeared but when we study the few remaining fragments of archaic terracotta revetments, the archaic reproduction in stone of what were originally wooden features and, before all, the description of Vitruvius, we may unhesitatingly affirm that the triglyphs reproduce the ends of beams or rather the decorative grooved facing of the ends of such beams, secured in position by pins or pegs passing passing through the projecting taenia or fascia surmounting the architrave and through the regula or short strip under the taenia, below each triglyph; the pins became the trunnels or guttae, still detached from the architrave as if driven in diagonally from below. The mutules or projecting blocks on the soffit of the cornice are as clearly the ends of the rafters of the roof, likewise with pegs or guttae, and all the other details are easily interpreted as translations of wood or terracotta members into stone, the metopes being the terracotta facing of the brick walls between the triglyph beam ends. Whilst the mutules and their interspaces always continued to represent the approximate slope of the roof in the peristylar temple, the triglyphs gradually came to be employed only in a decorative sense, as they did not correspond to the cross-beams of the peristyle ceiling, which were at a much higher level. This, however, was merely the result of the translation of the entablature into stone; the primitive wooden form was probably quite consistent, the beams coinciding with the triglyphs; and even in later stone architecture we have examples of such coincidence, as in the eastern portico of the Propylaea of the Athenian Acropolis.
From a constructional point of view it is quite obvious that the triglyphs are set in the axis of the columns, being the ends of the cross-beams that supported the ceiling and the roof of the temple. Any other position would have weakened the construction.
But the Greeks who loved order and good proportion, especially between the measurments of triglyphs and metopes, had a difficulty with this distribution of the triglyphs when they came at the corner of the building. Putting the triglyph above the center of the column there remained an empty space at the corner of half a metope. This empty corner was against the aesthetics of the Greeks. They solved the problem by shifting the angle columns inwards half a metope. So the triglyph could be placed at the corner and the metope could keep his proportions. The figure below is the front portico of the Parthenon at Athens. The shift of the angle columns is clear.
Vitruvius, in period 2 of this chapter, mentions this practice. But in the same time he is criticizing this shifting as inappropriate. It is clear that the Romans preferred an apparent system in the division of the architecural elements, in contrast to the Greeks who were clearly more flexible and adopted subtle solutions that were hardly seen at first sight.
When we analyse the text we can see some remarkable statements of Vitruvius. First we have his story about Hermogenes who decided to not build a Doric temple but build an Ionic temple instead. This story known in literature as the 'Hermogenes anecdote' is hardly believable. Hermogenes lived in the 2nd century B.C. while the Doric already found its decline in the 4th century B.C. Nevertheless Vitruvius considers Hermogenes as the inventor of the 'eustyle' Ionic (see III,3) and the statement at the beginning of this chapter seems to confirm this opinion. Further it is very unlikely that an architect could change the plans of a building under construction on his own initiative. But the reason why - according to Vitruvius - Hermogenes rejected the Doric style, fits completely in the Vitruvian system which tries to explain all temple architecture from a fixed module from which each part of the temple can be explained. This system posed a lot of problems in the triglyphs and metopes of the classical Doric temple as compared with the Vitruvian Doric temple.
Indeed, when we look closely at the Doric temples of Greece and Asia Minor, we can hardly find any common 'system' of proportions. Each temple seems to follow his own rules. A common factor can be the placing of the triglyphs above the centre of the columns and above the centre of the intercolumniation. Exception was made for the corner intercolumniation as we explained above; the distance between the corner triglyph and the triglyph above the next column was divided in two equal portions and the triglyph which was normally placed above the centre of the architrave between this intercolumniation was now shifted slightly to the corner. The result was two almost square metopes while the other metopes were square or nearly square as Vitruvius points out in sentence 2 of this chapter; each intercolumniation had two metopes.
Greek Doric order
Vitruvian Doric order
The Vitruvian Doric temple differs completely from this Greek layout. In accordance with his system he discerns - as he already did in the Ionic temple - diastyle (intercolumniation = 3 x lower diameter of column) and systyle (intercolumniation = 2 x lower diameter of column) temples with a further subdivision in tetrastyle (4 columns in front) and hexastyle (6 columns in front). The module is fixed by a division in equal parts of the length of the stylobate and that division is function of the type of temple one wishes to build; this gives following result: diastyle - tetrastyle: 27 parts, hexastyle 42 parts; systyle - tetrastyle: 19,5 parts, hexastyle: 29 parts. From this module all the other parts of the temple are deduced, a practice unknown in Greek - and even hellenistic - architecture.
Further Vitruvius avoids the problem of the corner contraction by shifting the triglyph above the corner column to the center of this column. This resulted in an open space of half a metope on the corner of the trabeation.
Another deviant feature of the Vitruvian doric is the thickness of the architrave which is fixed to one module. This is in strong contrast to Greek architecture where the thickness of the architrave didn't depend from a module and approximated the upper diameter of the column.
When we look at these texts and compare them with teh remains of Greek, Hellenistic and Roman architecture, it is clear that Vitruvius has written a theoretical approach of temple architecture. The module in Ionic and Doric style gives a common measure and when it comes to Corinthian the trabeations are interchangeable.
When Vitruvius still mentions the Greek manner, the other authors intentionally ommit this Greek way of building; they only work with the Roman point of view as described by Vitruvius in this chapter.
In Vitruvius' text Doric columns have no base at all, and this completely in accordance with ancient building practice: in Greek and Roman Doric order we find no bases under the columns as a transitional element between stylobate and column. During renaissance and later builders felt the necessity to insert a base. This base was copied from the so called Attic base of the Ionian order, described by Vitruvius in III,5,2. The reason for this insertion is given by Alberti, VII,7: It is essential in all building, as we said, to take care that everything rests on a solid base. In the reasoning of renaissance builders this solid base was only provided by broader element: a base fashioned in the Attic manner.
Les dix livres d'architecture de Vitruve, Corrigés et traduits en 1684 par C. Perrault, Paris, 1684.
Vitruvius, De Architectura libri X, ed. F. Granger, London, 1962.
Ton Peters, Vitruvius, Handboek bouwkunde, Amsterdam, 1999.
W.B.Dinsmoor, The Architecture of Ancient Greece, London, 1950.
H.Knell, Die Hermogenes-Anekdote und das Ende des dorischen Ringhallentempels, in Vitruv-Kolloquium des Deutschen Archäologen-Verbandes, Darmstadt, 1984, pp. 41-64
B.Wesenberg, Vitruvs Griechischer Tempel, ib., pp.65-96
H.Knell, Vitruvs Architektursystem, Darmstadt, 1985
J.J.Coulton, Modules and Measurements in Ancient Design and Modern Scholarship, in Munus non ingratum, Leiden, 1989, pp.85-89
B.Wesenberg, Die Bedeutung des Modulus in der vitruvianischen Tempelarchitektur, in Le Projet de Vitruve, objet, destinataires et réception du De Architectura, Rome, 1994, pp.91-104
Philibert De l'Orme, Architecture, Rouen, 1648
Andrea Palladio, The Four Books of Architecture, London, 1738
J.Ryckwert - N.Leach - R.Tavernor, Leon Battista Alberti, On the Art of Building in Ten Books, Cambridge, 1988
V.Hart-P.Hicks, Sebastiano Serlio, On Architecture, New Haven-London, 1996
B.Mitrovic, Giacomo Barozzi da Vignola, Canon of the Five Orders of Architecture, New York, 1999
Back to contents Book IV
Suggestions and remarks?
don't hesitate to send me a message