The choice of the site for the excavations of 1905 was influenced by the interest arising from the discovery of the cathedral of Saint Gregory, or Zvartnots - the "Vigilant Forces", by the archimandrite Khatchik in the locality of Arapar, not far from Vagharchapat. Excavations that began in the year 1900 uncovered this cathedral, and for a while it became an object of detailed attention from academics. The interest in the structure is connected to cultural questions concerning Grecophile Armenians of the 7th century. However, the discoveries revealed by the excavations were incomplete and some important aspects remained uncertain. It was therefore natural to wish to fill those gaps by excavations at Ani, where we knew the cathedral at Arapar had been reproduced by the Armenian king Gagik the 1st.

The monumental cathedral of Gagik (site #15, fig. 67), now fallen into ruins, was the most important discovery revealed by our excavations of 1905 and 1906 ¹⁹². The construction at Ani, during the reign of Gagik 1st, of this splendid cathedral of Saint Gregory, based on the plan of the 7th century church built at Vagharchapat, is related by the historians Assoghik, Kirakos, Mekhitar Ayrivanetsi and Samuel Anetsi. It is one of the rare cases where Armenian historians have recorded relatively exact information on a building at Ani. All the authors mentioned speak in amazed terms about this work of king Gagik. From the time of my first travels to Ani, I had reflected on the probable location of the site of this cathedral; different sites had been considered, but we lacked the exact data that would have allowed us to choose one site with certainty. The incorrect identification of the structure of king Gagik with the small family church of St. Gregory, still standing, and belonging to the house of Aboughamrents, comes from Brosset ¹⁹³. Lynch ¹⁹⁴, who saw Ani in 1894, remarked that except for its site above the Tzaghkotsatdzor (Valley of the Flowers) all the evidence was against this identification. Excavations by the archimandrite Khatchik around Vagharchapat gave us the information that allowed us to establish more exactly the location of the ruins of Gagik's cathedral. Thanks to this, Vruyr, whose free time was devoted to photographing the antiquities of Ani, was the first to identify the hill above the Tzaghkotsatdzor where the vestiges of the church based on the building of Nerses were to be found. This huge hill (fig. 68), which had the remains of foundations at two places, had attracted my attention in 1892, and it had been set aside for future excavations. In 1894, Lynch was interested in the same hill and included in his notes the information of an old priest, who said the place had once been the house of the synod ¹⁹⁵.

Seen from the east, the site stood out because of some rows of isolated foundations, resembling the vestiges of a wall, but which actually was, as we discovered later, the remains of one of the four additional pillars of the church, specifically the one behind the altar. The quantity of foundations increased beside the slope of a high hill that was used as a place to graze cattle. It was here that we began our excavations in 1905, ending only in 1906. A good part of our time and effort was devoted to excavations to the south-west and west of the hill. To the west, a street was uncovered. To the south-west, we discovered the habitations of poor people. After doing this, all our efforts were focused on the heights containing the ruins of the cathedral. Our Armenian and occasionally Turkish labourers had just begun to extend the excavations into the second layer of the hill when we discovered the voluted capital of one of the columns (fig. 69).

The Upper Layer

After excavating the hill to the ground level, and the exterior of the cathedral to its foundations, we saw that this artificial hill, which had a height of 8.75 metres at the middle, had formed gradually. Within its relatively small depth we discovered three different cultural deposits representing three periods of the brief but active life of Ani, starting from the early years of the 11th century. The hill was thus divided into three layers. The upper layer, with a depth of 25 to 85 centimetres, contained a cemetery of the most recent period and the last to develop. Amongst the multitude of graves excavated in 1905, only one contained a wooden coffin, roughly nailed together. The dead were buried completely naked indicating that this cemetery had clearly belong to an extremely poor section of the population. These burials were certainly those of Armenians of Ani during the period of its decline.

The Middle Layer

The middle layer had a varied thickness: it reached 6 metres at the middle of the hill, but its depth decreased at the slopes. The middle layer also contained a cemetery of isolated graves, but of a higher quality. Wooden coffins were not very numerous, but they were better finished and were held together using iron clamps. We noticed a greater care given towards the dead: they were laid into tombs of stone slabs carefully arranged in the shape of coffins. Outside the cathedral, this type of burial was observed much lower down, as we saw the four-sided openings of the tombs within the cross-section of the earth deposit, still intact in 1905, near to the walled-up north entrance of the church that we had excavated to its foundations. In addition, even in this middle layer, in places tombstones were found in the form of flat slabs with double slopes in the shape of a cradle. A lone funerary monument, uncovered to the north of the statue of Gagik and 9 metres from the church, was of a relatively complex construction, although from a more recent period. Its pedestal was composed of three superimposed slabs. On the pedestal, a plinth held a khatchkar carved with wonderfully beautiful decoration: at the base of the cross a carved shield was decorated at its edge with motifs of pomegranates and clusters of grapes (fig. 70). Only fragments of this sculptured cross with scraps of an Armenian inscription were discovered. The same layer also contained the ruins of poor people's homes.

The Lowest Layer

The content of the lowest layer dates from the oldest period of the culture of Ani. This layer contained the remaining in-situ vestiges of the cathedral and the remains of its initial destruction. The height of the lowest layer varied from between 1.55 metres to 2 metres ¹⁹⁶. The excavations of 1905 reached the ground level, but not over the entire area of this sector, of which only a little more than a quarter was excavated. The elevation of the altar of the church (fig. 71) was revealed only on one side, mostly around the section with the south staircase near the south-east pillar.

The south colonnade was completely uncovered (fig. 72). The photograph (fig. 73) shows the inaccurately placed capital on the lower cylindrical section of a column, which represents only a third of the original height of its shaft. It had been placed there during the excavations, when the surrounding earth had still to be removed. We can get an idea of the dimensions and the weight of the different sections of it according to figure 74, which shows workers hoisting a section of a drum onto the base of a column. On the south side, a gallery was exposed; it separated the external wall from the pilasters of the cruciform central section of the church. An isolated column with an excessively massive foundation, a later construction, constricts the freedom of movement in the gallery. During 1905, the south entrance (fig. 75) was excavated from the earth; it appears to have been the only one not to have been walled up, although it also was encumbered by a later pillar, partly visible through the opening of the entrance.

Excavations showed that the destruction of the church has been gradual and that, at some time, measures had been taken to consolidate the structure, which had clearly suffered from an earthquake. At a certain period, fundamental repairs had been undertaken: these had demanded considerable revisions inside the church, to the extent of modifying some basic details of its plan. When it was realised that the collapse of the cathedral had become inevitable it was emptied in advance of all its precious objects, priestly clothes, manuscripts, sacred vessels, etc.

Excavation of the lower layer uncovered fragments of the structure, sometimes huge, that had collapsed as whole blocks during the initial catastrophe. Some parts of the church, especially the pillars, had remained upright for a long period. It was against these pillars that the residents of Ani leaned their houses built on the surface deposit that covered the structure and in which they later placed their burials. At the same time, these residents pulled off the huge facing stones from the walls and the surviving pillars or removed the massive cylinders of the columns in order to reuse them in various ways.

The huge difference between the cathedral and the later constructions is easy to distinguish in figure 76. To left can be seen the remains of later houses, and on the right can be seen the southern part of the cathedral starting with the south entrance, looking towards the west. The old foundations have a stepped base of three levels running all around the circumference of the church. On the upper level rest the bases of 36 twin half-colonnettes which had divided the walls of the lower facade into many faces that were crowned at the top, below a line of windows, with a blind arcade (fig. 84).

Taking an overall view of the ruins discovered of the cathedral, it is the north-east side that is the most damaged; we can see the row of half-colonnettes that decorate the exterior of the lower facade (fig. 77). On the other sides, these half-colonnettes are preserved to a greater height. This view allows us to glimpse inside and see the columns and the remains of the original piers, including those of the southwest, the northwest and the northeast corners, as well as the additional pillars, including those of the western, northern, and, in the foreground to the left, eastern ends. The three levels forming the stepped base of the church are reduced to two only at the location of the entrances, where the third level is cut through: for an example of this see the opening of the south entrance (fig. 75). But the two lower levels, which are far too high to serve as steps, remain in place and we have not succeeded in discovering how the believers managed to enter the church: there is no trace of a staircase. Through the south entrance can be seen the remains of the southwest pier, which was hidden under the earth before the excavations, and the south pillar, a later construction inside which is visible, in its collapsed upper section, one of the original columns with its capital that was later hidden inside the pillar (fig. 73).

The Interior of the
Cathedral of Gagik

There were walls in the interior of the cathedral, consisting of crude foundations from a later period, that fill the space between the semicircular colonnade of columns as well as the narrow gallery passage between the north, east, and north-east pillars and the corresponding parts of the external wall. We initially believed that at some period these foundations had served to support the church as it threatened to collapse. But in the base of these walls, between the northeast pier and the external wall, there is a column drum (fig. 78) in a transverse position, and between the north pillar and the external wall there is another cylinder and a capital in a horizontal position. Therefore, the cathedral had already been at least partially destroyed by this period. In addition, the line of a similar wall starts at the southwest pier and travels in a diagonal line to the northern edge of the altar elevation, thus dividing the central part of the church into two unequal halves (fig. 79). Opposite the middle part of this later wall, close to the elevation of the altar, excavations uncovered a fragment of the polygonal drum: this discovery definitively solved the problem of the form of the dome of the church. These huge blocks, discovered to be from the collapsed upper levels, were a priceless aid towards understanding the more complicated transitions in the vertical structure of the cathedral. Important finds included a huge fragment discovered just to the west of the entrance on the north side: this was a part of the hidden passage with the start of its longitudinal vault forming the circumference of the church.

In two plans, the cathedral has been reconstituted in all its details. The plan of the most recent period (fig. 80), showing the essential repairs, has in its centre a circle with the same diameter as that of the interior circumference of the polygonal drum, which was circular inside. The base of this drum was the usual square set on four arches, their bases resting on four lofty columns joined to the four piers.

The middle circle has the same diameter as that of the circumference of the second level of the church. It was here that the repairs were especially concentrated, as well as on the arches supporting the wall of the second level and the columns on which these arches rested, eight in number, following the line of the middle circle.

The eight arches rested on the middle pair of columns situated on the curve of the four semicircular colonnades, and on the four isolated columns opposite each of the four original piers. However, the base of the wall of the second level was not immediately set onto these eight arches that rose from each isolated column to descend onto the nearest pair of middle columns. The wall of the second level, also visible on the exterior, rose towards the upper parts (fig. 81) of the structure, forming the interior wall and separating the gallery from the central space inside the church. This gallery, with a width of only three metres, ran around the church without interruption, bypassing the altar elevation from behind. It widened at the corners of the cruciform central space, at the angles of the junction between the semicircular colonnade and the piers where, in the angles of the upper section, there clearly had to have been small rooms. This gallery occupied the full height of the lower level (fig. 82). Over the length of all the circumference, it was divided in two levels: a vaulted upper gallery comprising a secret passage to four angular rooms, and an lower passage, also vaulted, open to the middle of the church up to the height of the colonnades with their semicircular arches. Windows within the secret upper passage looked out on two sides: oblong windows opened onto the interior of the cathedral and porthole windows opened onto the exterior. The longitudinal vault of this secret upper passage linked the external wall of the first level with the internal wall that then continued upwards as a single wall acting as both external and internal wall. The longitudinal vault of the passage inside the gallery, half open to the central space, also joined the external wall of the lower floor level to the same internal wall which, consequently, supported the whole of the external wall of the second level and half of the two vaults coming from the peripheral vault of the cathedral, that is to say from the vault of the secret upper passage of the gallery and the vault of the lower passage that communicated with the central space via the openings between the columns. It is this internal wall, put under triple pressure, that rested on the eight large arches which pressed their whole weight onto the twin central shafts of the semicircular colonnades and onto the isolated columns next to the piers.

For the moment, I will remain silent on why the columns could not support this weight, and why the cathedral had been shaken and covered in cracks of which those in the southwest section were clearly the most threatening. Essential repairs had been undertaken: the central pair of columns in each of the four colonnades had been encased within newly constructed east, south, west, and north pillars. One of the eight arches, which supported an excessive weight and must have contained the most dangerous crack, had been consolidated by a new set of supports (fig. 83); for it had been necessary to construct an additional arch which spanned from the two middle shafts of the west colonnade to the isolated column near the southwest pier. This isolated column had been strengthened by additional foundations. These are the main characteristics of the essential repairs. In addition, as mentioned earlier, the west and north entrances had been walled up because the newly constructed pillars were bulky and almost barred the passage to the centre of the cathedral.

The southwestern isolated column has preserved the remains of the lower foundations of the pillar that had encased all of its shaft and capital after the repair works. The west pillar (fig. 72) dates from the same period: in the collapsed foundations appear two columns, the central pair of the west colonnade.

The project to reconstruct the external facade of the west side (fig. 84), done by the architect Toramanian, reveals its internal structure in a vertical section (fig. 85), and the west colonnade is depicted in its two states: the original (fig. 82) and the consolidated (fig. 83).

Although the design had not undergone fundamental changes, its internal aspect before the repairs was one of greater clarity and spaciousness (fig. 85). Nothing obscured the central space comprising an equal armed cross with semicircular ends; each arm inscribing a clear semicircle with its six columns, the open space between the two central columns revealing a view of the route to the middle of the cathedral to those entering by the doors. The east arm was entirely occupied by the elevation of the altar, which encroached a little into the central space. One climbed up to the altar from the north and south arms; on each side there was a staircase with five steps. These staircases were joined to the piers (fig. 71). The facade of the altar elevation was flanked, to the north and to the south, by a decorative blind arcade resting on twin half-colonnettes, surmounted by a cornice similarly decorated. The decoration of this cornice was an interlace like that of the cornices on the first and second levels but on a smaller scale. The elevation of the altar was shielded by a screen of two or three rows of masonry that rose from the same dais of the altar, along the semicircle of the eastern arm. It was at this level, on this screen of the altar elevation, that the six columns of the eastern half-circle were located (fig. 82). The original height of the column shafts is known from the column preserved in situ near the northwest pillar (fig. 86). The start of the arch connecting the pier to the upright column on its left survives, high up on the pier. The other side of this arch rested on the capital of the column. The capitals of the columns in the semicircles are, in general, all of the same form although, in details, none are identical to each other and each has some difference in its decoration (figs. 69, 78, 87). The arches from these columns not only connected the piers to the capital of the nearest column, but also connected all the columns in the semicircular colonnade, capital to capital. Excavations recovered a sufficient number of capitals and column cylinders; pendentives recovered were equally numerous. These last were located at the angles of the gallery vaults, formed by the junction of the colonnade and pier. The capitals of the isolated columns were of a distinctive and original shape, without any sort of decoration (fig. 78).

The External Decoration
of the Cathedral

Decoration on the cathedral was more abundant on its exterior. The ornamentation of the doorways had been made a particular object of attention. Their pediments, along with their frames in general, exhibited massive features when compared to their low and rather narrow openings. Carved fragments of the pediment over the west entrance had been discovered during the excavations of 1905. These included a huge stone with an acanthus leaf motif, and the remains of an equally large stone decorated in the local style: a fine work, almost lace-like. Decorations on the south entrance, also discovered in 1905, were of an extraordinary beauty. However, unfortunately they had been removed: we succeeded in finding only a few small examples. We also found fragments of the decoration of the pediment over the north doorway. Acanthus also formed part of the decoration of the side entrances, but they were smaller and more finely worked.

A series of discoveries gave some idea of the numerous ornaments that decorated the exterior of the cathedral. We have, among others, a banded ornament on the lower section that ran all around the cathedral above the blind arcade; another banded ornament that encircled the drum; ornaments coming from the frames of windows, from the blind arcade, etc. Ornaments from the blind arcade on the second level were individual in that each arch had its own design. The decoration of one of the two cornices, a half-rosette interlace (fig. 88), should be specially pointed out, so also should the ornament on the blind arcade that crowned the pairs of half-colonnettes on the exterior of the lower level. This decoration represents a garland of stylised vine leaves ending in an arrow on the archivolts (fig. 89). This ornament ran all around the cathedral, repeated on all 32 arches of the blind arcade.

The dimensions of the facing stones on the foundations of Gagik's cathedral are particularly deserving of attention, since the historian Stepanos Taronatsi in his description of this church ¹⁹⁷ described them, with legitimate astonishment, as huge carved blocks similar to rocks.

I hoped to be able to observe the influences of Byzantium represented in this cathedral of Gagik, introduced into 7th century Armenian culture by catholicos Nerses III. In reality, it is a result of the "Sturm and Drang" cultural ethos of the 7th century, introduced into the architecture of the country by the Armenian Chalcedonian movement, at the head of which was Nerses III.

The Statue of King Gagik

Undoubtedly, the most important decorative object, more due to its form and its workmanship than its importance in the external decoration of the cathedral, is the statue of the builder of the church, king Gagik the 1st (fig. 90).

This precious discovery was unexpected because nobody imagined that the ancient Armenians had developed a real art of free-standing sculpture, beyond the needs of ornamentation. In addition, historians of Christian art in general deny the existence of sculptural work within Oriental Christian art after the 8th century. They give the following reasons: the first being the iconoclastic period and the second being Islam. Yet a statue is discovered here in Armenia, where iconoclasm was particularly strict and the influence of Islam doubly strong. We do not need to be especially shrewd to see, on examining its characteristics, that this statue is not the first attempt of an artist and that, even with its imperfections, the work bears the stamp of coming from a school with a well established tradition. The statue was placed high up, in a niche on the north wall of the cathedral, the king holding a model of the church in his hands. Unfortunately, only fragments of the lower level of this model survives. The statue and the model of the church had been discovered in fragments and their skilful assemblage was the work of the artist Poltoratzki, who was also the author of the realistically coloured depiction of the statue holding the model of the church. The artist has recreated only the upper part of the model (fig. 91).

The statue of Gagik was carved from a pinkish stone block of large dimensions, since it reaches a height of 2.26 metres. The statue was polychrome: the clothes and the face were coloured red, the beard and moustache were black, the turban was white with a red band above the front; the unusual draping sleeves were also white.

An Armenian Inscription

Beside the statue, on the north wall of the cathedral and at an original height of approximately 10 to 11 metres, was found an Armenian inscription (fig. 93) in large and beautiful characters. We succeeded in discovering only a part of the inscription, finding 131 fragments in total. The last three lines of the section which we were able to reassemble reads: "on the order of the Shahanshah Gagik, son of Ashot, king of the kings of Armenia. Witness to God and catholicos Sarkis".

The condition of this inscription illustrates not only the way in which inscriptions have been broken and lost during the collapse of structures, as well as the amount of effort that is demanded from the seeker and the ephigraphist to uncover, re-assemble, and reconstruct these precious documents of civilisation, but above all the way in which the destructive forces of History have broken and annihilated the existence of ancient Armenia.

The Chandelier

The discovery of the chandelier was of great importance (fig. 94), regardless of its bad condition. This chandelier is of the crown type of chandelier. The crown is a circle of 68.9cm diameter that consists of a decorative belt 13.25cm high, engraved alternately with figures of eagles with outstretched wings and unidentifiable winged monsters inscribed inside 12 circles. Holders, intended to receive small glass containers, are fixed around this crown, on two levels. The thirty holders of the upper level are soldered onto the upper rim of the crown, while the eight holders of the lower level are soldered to the lower rim of a belt around the chandelier. These holders consist of small metal disks attached to short sleeves, also of metal. There is a circular opening in each disk to receive a small glass container, whose long pointed tip would have ended in a spherical bulge. The simple holders of the lower row alternate with representations of flying doves, eight in number, carved out of copper sheet. These doves seem to fly around the crown with their wings spread horizontally. The two wings and tail of each dove have circular openings intended for glass containers. In addition, each dove holds in its beak the metallic thread of a triple chain, that is to say three threads, each consisting of two strands. A small, flat-bottomed lamp hangs by three lobes from this triple chain. Thus, each dove held four lamps. The crown, with all its supports, and copper doves, and all its containers and glass lamps, is fixed to six metallic ribbons that are joined at the top, next to a buckle that served to suspend the chandelier from an iron chain which had a length of at least 19.2 metres and was formed from 192 links ²⁴². Isolated holders are seen here and there on the metallic ribbons. At the top, at the junction of the metallic ribbons, a straight stem of copper leaf with two pairs of holders descends inside the chandelier. A miniature chandelier is suspended from this stem. It is actually a flat circle hung from four metallic ribbons, which has four miniature holders intended to receive four miniature containers. This straight stem continues down even lower and a small metallic basket, crafted with very fine perforations, is hung there by a well-worked chain. This basket was intended to receive a large glass lamp. We also found brackets - miniature doves like those previously described - whose location was probably in the perforated basket. The glass containers and lamps were of different colours: white, yellow, turquoise, and of various dimensions. It is probable that when the chandelier was suspended at a specific height and lit, some lamps shone out while others scintillated. The chandelier could receive more than 114 lamps and glass containers in total. From all the material discovered, the artist Paltoratzki created a coloured drawing, with the broken or damaged parts reconstituted.

Deprived Housing

The excavation of the cathedral of Gagik showed that after its destruction, the vacant space around the ruins had been used for private habitations, some partially joined to the cathedral, in which lived, based on our discoveries, Muslim Turks or Persians as well as Christian Armenians. From all viewpoints these habitations are crudely constructed. They do not follow a standard plan, because the disposition of rooms and their shapes are haphazardly arranged.

Sometimes, the rooms are almost triangular, always with asymmetrical foundations, for the most part badly constructed; if they have not been removed from the old ruins, the stones are roughly cut. Nevertheless, these humble hovels are remains dating at a minimum from the 13th and 14th centuries. The people that constructed them had no respect for the ruins of the cathedral or those of neighbouring structures, from which they took polished facing stones and decorative elements to reuse in foundations or as lintels and beams. These new builders demonstrated the same barbarous indifference for the monuments in the cathedral courtyard. They broke and removed whatever they could, including the ornamented parts of funerary monuments by mutilating the pedestals; or alternatively, if the work was heavier than their capabilities, they used the monuments as links on each side of the miserable walls of their sad hovels.

The surface around the cathedral had been excavated on each side to a distance of ten metres or more from the external wall. To the southeast, excavations uncovered architectural fragments and ornamented khatchkars that continue up to the foot of the hill of a certain "miserable Gregory" (site #16). This hill hides, in all probability, the ruins of a mausoleum chapel. On the open part of the site, to the southeast of the cathedral, there was a large mortuary monument, of great age, of which survives only as a pedestal of two steps (fig. 95) surmounted by a plinth with mutilated mouldings. We do not know if this pedestal supported a stele or a khatchkar ²⁴³; the lower parts of the monument, still in-situ, served as the foundations of a later construction.

To the east, a wall joins that of the cathedral at an angle. It is a part of a rectangular oblong chapel that was joined to the cathedral and communicated with it via the gallery behind the altar (fig. 96). If judged by the plan of the cathedral of the same type excavated near Etchmiadzin by the archimandrite Khatchik, this rectangular structure that is joined to the eastern end of the cathedral has to be contemporary with it. Nevertheless, after careful examination we became convinced that at Ani the chapel was joined to the eastern end of an already completed cathedral. The chapel's base has only two steps.

The site where the monumental round cathedral of Gagik was erected is encroached on the southwest by another large building (site #17).

The evidence unearthed by excavation showed that this structure was a reconstruction of the 12th or the 13th century. Its facade, decorated with a mosaic of red stars and sculpted black rhombi and other remarkably rich ornaments (figs. 97, 98) is depicted here in a reconstitution by the architect Toramanian (fig. 99). This elegant facade testifies to the high standard of decorative techniques that developed in Ani's civil architecture during the period previously mentioned, that is to say the 12th and 13th centuries. The narthex of the church of the Holy Apostles (figs. 100, 101) also dates from this period, and is covered in sculpted ornament that runs up the height of the external facade. In addition, the interior of this church is even more remarkable with its decorated mosaic ceiling composed of stones of various colours and different configurations.

An examination of the facade of this large building revealed more than 43 motifs. One of the ornamented stars of the sculptured mosaic bore the inscription "Sarkis" in Armenian (fig. 102); this would perhaps be the name of the artist or architect rather than the proprietor. Or perhaps the structure served as the residence of the archbishop of Ani, Sarkis the 1st (1209-1211), or Sarkis the 2nd (1245-1276).

NOTES:

192. Excavations and Works at Ani in the Summer of 1905 (preliminary report), Report 3, Vol. X, Saint Petersburg, 1907. See also Report of the Archaeological Commission for the year 1905, p.75-76, and for the year 1906, p.105 (in Russian).
193. The Ruins of Ani, vol. II, p.105, note 4. On the Aboughamrents church see Alishan, p.51-52.
194. Lynch, Armenia, vol. I, p.382.
195. Op. cit., vol.I, p.384.
196. There are some indications that, in the part between the colonnade and the external wall of the cathedral, the ground is not virgin soil.
197. Stepanos Taronatsi, p.282.