Introduction

1Since the early 20^th century, Haghia Triada has represented a major source of interest for the study of the Cretan Late Bronze Age, thanks to its outstanding architectural record and variety of movable items, dated to the Neopalatial, Final Palatial and Postpalatial periods respectively (LM I through LM IIIC in the local sequence; La Rosa 2010). The abundant evidence of Neopalatial date has been reported from time to time since the early 20^th century, resulting into a thorough excavation report in 1977 (Halbherr et alii 1977); conversely, the evidence dated to the LM II through LM IIIC phases could only be properly evaluated after Vincenzo La Rosa began a new cycle of fieldwork at the site in 1977 (La Rosa 1977; 1979-80; 2010; Militello 1998; D’Agata 1999).

2Concerning the Final Palatial period, scholars of Minoan architecture have particularly emphasised the importance of Haghia Triada’s LM IIIA2-B monumental buildings within the framework of contemporary Crete. Along with those recently excavated at the not distant harbour town at Kommos, such buildings have long fuelled the debate on the ‘Mycenaeanising’ character of the island’s architecture in the aftermath of the fall of the second palaces (Hayden 1981; 1987; McEnroe 1979; 2010: 128-132; Darcque 1990; Driessen & Farnoux 1997). Accordingly, their layout, as well as technical features such as the masonry, door jambs and stone thresholds have been taken into account from time to time, enabling scholars to conclude that they “represent innovations without parallels in Crete, far different from the rest of the island” (Shaw & Shaw 2006: 874), but somewhat matching their counterparts on the Greek mainland (Hayden 1987; Cucuzza 1997). Yet, the debate on the local architecture has been limited to a few major complexes, such as Megaron ABCD and Stoa dell’Agora at Haghia Triada and Building P at Kommos. At the former site, the recent publication of a group of LM III buildings by one of the authors of this paper (Privitera 2015a), located in an area that was christened the Villaggio by the early excavators, will hopefully enable new meaningful evidence to be involved in the ongoing debate. In fact, that publication is essentially descriptive in character, being concerned, on the one hand, with retrieving from the unpublished archival documentation (including notebooks, sketches, and photographs) as many data as possible on the early excavations at Haghia Triada and, on the other, with presenting in detail the extant structures and finds in order to make sense of architectural forms and functions.

3Against such a backdrop, the present paper intends to supplement the available description of both structures and finds from the LM III Villaggio by focusing on the most monumental building located within its borders, the Casa dei Vani Aggiunti Progressivamente (Privitera 2015a: 35-83), in the light of data deriving from the application of space syntax analysis (Hillier 1996; 2014). The idea of selecting such a building as a case-study essentially draws on Letesson’s research concerning Neopalatial architecture (Letesson 2009; 2014), which conclusively showed how the arrangement and inner organisation of about seventy Minoan buildings betray the existence of a subjacent architectural set of principles or a ‘genotype’. It is our contention that employing such a methodology makes it possible to raise several interpretative questions, such as to what extent Casa VAP shows formal links with the Neopalatial building tradition and whether the cultural code/s that inspired its layout evolved over its long period of life (ca. 150-175years, from early LM IIIA2 through late LM IIIB). It is worth stressing that this paper could not have been written, had its authors not met on the occasion of a cycle of Aegean conferences that Jan Driessen organised in the Fall 2015 at the Université catholique de Louvain. Therefore, it is a great pleasure for us both to thank him for inviting us to write this paper. It represents but a small token of gratitude to the Belgian Genius Loci of Aegean archaeology, Robert Laffineur, as a way to acknowledge the distinctively incisive role he has played by bringing together so many scholars of different academic background and nationality at the many Tables rondes egéennes he has organised for the past thirty years.

1. The LM IIIA2-B Villaggio at Haghia Triada and the Casa dei Vani Aggiunti Progressivamente

4The LM III Villaggio at Haghia Triada corresponds with the north sector of the site and more precisely with the slope located to the west of the court of the Agora (Puglisi 2007; 169; Privitera 2015a: 5-10) (Fig. 11.1). This area represents nearly one third of the settlement but, in contrast to the imposing buildings located to the south, has remained out of focus in scholarly opinion up until very recently (Puglisi 2007; 2014; Privitera 2011; 2014; 2015a). After the LM IB generalised destruction at the site, the earliest traces of a reoccupation in the Villaggio are LM II in date, even if the first roofed spaces were built as late as LM IIIA1 (Privitera 2015a: 134-135). Such structures did not survive into LM IIIA2, during which the area was substantially reorganised. In particular, a ca. 46.50 m long LM IA retaining wall running east-west, christened Muraglione a Denti (Puglisi 2007; Privitera 2015a: 7), was reused as the north wall of a large residential building, Casa dei Vani Aggiunti Progressivamente, which occupied an area corresponding with the southern half of the Villaggio. To the north of this wall, the slope was divided into three major terraces and was occupied by buildings mostly devoted to the storage of bulk staples (Casa delle Camere Decapitate, Edificio Nord-Ovest, Edificio Ovest: Privitera 2014; 2015a: 97-122) and by a group of smaller auxiliary spaces (buildings X, Y, and Z: Privitera 2015a: 85-95).

5The Casa dei Vani Aggiunti Progressivamente (literally, House of the Progressively Added Rooms, henceforth Casa VAP) (Fig. 11.2) is one of the largest buildings constructed at Haghia Triada in LM III. The odd name is due to the way in which it was progressively enlarged by means of single rooms, a process which involved the transformation of original exterior walls into partition walls. Together with the long precinct which adjoins it to the south, christened Edificio W, and a pair of small annexes to the east of the latter, labelled E and F respectively, it consists of 17 ground floor spaces and occupies a roughly rectangular area of ca. 26.5 m east-west by ca. 28 m north-south. The combined expanse of all these spaces comes to ca. 450 m².

6Casa VAP has the most complicated building history by far in LM III Haghia Triada (Fig. 11.3). Even if it has only produced a limited amount of LM III pottery, the outline of its building sequence is relatively well known; its lifetime is hinted at by the discovery of LM IIIA2 and LM IIIB pottery at various spots, both indoors and in the outdoor adjacent spaces. The building sits on top of a few Neopalatial buildings, which were all destroyed by fire in LM IB, and is structurally later than a short-lived LM IIIA1 building of probably ritual character (Edificio E; Privitera 2015a: 29-33). Accordingly, its very first nucleus was built between a very late stage of LM IIIA1 and the beginning of LM IIIA2 (Privitera 2015a: 75-78). By this time, its overall outline corresponded with the area encompassing spaces G/H/I, L/M, and N, that is, a rectangle ca. 16.90 m east-west by 7.75/8.25 m north-south. However, the inner layout differed a lot from the current one: G/H/I originally was an undivided space, ca. 2.90 m east-west by 6.20 north-south, endowed with both a central entranceway and an elegant façade dressed with ashlars to the south. It communicated with L/M by means of a passage placed on its west wall, which was blocked when space H was created. Conversely, a further opening placed next to the north-west corner of G quite probably did not exist at this stage, being created later, in order to make space G accessible from space L. Also L/M was a single space, since the inner partition between L and M is a much later construction, possibly part of a reoccupation of this part of the building in historical times. At this stage, the house probably was a one-storied building and safely had two, and probably three, main accesses from the south; its three inner spaces, moreover, communicated with each other through inner passages, one of which (that between H and L/M) placed next to two outer entranceways. Its area was ca. 130 m².

7In early LM IIIA2, a further rectangular room, space D, was built east of G/H/I and a new entranceway was opened in its east wall. At this stage, the creation of an upper storey is hinted at by the presence of a stair inside D, and quite probably by that inside I. Accordingly, despite occupying an only slightly larger area than earlier (ca. 160 m²), the indoor space of Casa VAP was substantially increased, as it is plausible to advance that an upper floor existed at least above spaces D, G/H/I and L/M, measuring no less than 90 m².

8Starting with this phase, the ground floor was spatially divided into two autonomous sectors. On the one hand, one could enter spaces G, L/M and N through the entrances in the south wall of both the latter (the passage between L/M and I was blocked, turning the latter space into a door-less space, possibly accessible from a trapdoor in the roof); on the other, the two entranceways placed in I and D provided communication with the upper floor, where the presence of the ‘domestic’ sector of the building can be envisaged (Privitera 2015a: 78-79).

9Still in LM IIIA2, spaces P/Q and A were added, representing a third building phase (Privitera 2015a: 79). The overall size of the building increased to ca. 330-360 m² (including the upper floor). P/Q are partitions of a large rectangular space, which probably contained one more wooden staircase inside Q. Such space was characterised by a high level of accessibility, being entered from outdoor through a monumental double entranceway on the south and another on the east. Both the south and east wall, dressed with elegant ashlars, attest to the importance of this space. From P/Q one could easily access the upper floor through both the stair in I and that in Q. Therefore, it seems plausible to advance that, like the ground floor, the upper floor was also divided into distinct suites of room. Moreover, the division of the ground floor into two self-contained sectors, which already characterised the second building phase, was emphasised by the blocking of the entranceway in the south wall of L/M, which later could only be accessed, along with G and N, through an entranceway that might have been located further west in the south wall of N but is unfortunately not preserved. Space A, more spacious than P/Q (ca. 4.50/4.75 m east-west by 6.50/6.90 m north-south; area ca. 42 m²), was endowed with an inner wooden pier standing on a stone base. Quite probably, due to its large size, it supported no upper floor. An entranceway placed next to the south-west comer functioned as one of the entrances of the building, along with that in P/Q. The south wall, moreover, had an elegant façade in ashlar masonry, which originally framed a window ca. 1.50 m across.

10In the fourth building phase, a small rectangular room, space C, was built east of P/Q and south of D. It remains uncertain whether it supported an upper floor or a terrace, both being accessed through the stairway placed in Q. Quite probably, space C originally represented a dead-end, like G and the south sector of space D, being used either as a storeroom or a bedroom. In fact, a low opening in the south wall of space D could have been either created upon building C or, more probably, at a later time, to enable the passage of commodities and people between C and D. It is a matter of speculation as to whether space O was built in this same phase. This was a large rectangular room, entered through an opening placed next to the north-west corner and endowed with a central pillar for holding the roof and low benches along the south and west walls. It did not communicate with the rest of the building and can be structurally considered one more dead-end space; however, its proximity to spaces L/M and N makes one wonder whether it was functionally related to the latter, thereby representing an addition to the west sector of the ground floor of the building. By the time of its abandonment, it contained fragments of storage jars and could accordingly function as a storeroom or a living space related to food consumption. By including both C and O in the same building phase, the surface of Casa VAP can be calculated at ca. 284 m² on the ground floor and possibly further 140 m² on the upper floor.

11A substantial rearrangement of the east sector of Casa VAP, dated to a mature stage of LM IIIA2, represents the fifth building phase (Privitera 2015a: 80). It is centred upon the refurbishment of space A with wall paintings that stylistically match the decoration of the celebrated Painted Sarcophagus excavated in the nearby cemetery (Paribeni 1908; Lang 1974; Militello 1998; Privitera 2008; 2015a: 62-70; 2015b) (Figs 11.4-11.7) and depict, among others, processions including musicians, altars, and animal sacrifices. On this occasion, space A was made higher than before, as the rising of the threshold of the south entranceway makes clear, and was turned into a darker room by blocking the large window in the south wall. Together with the wall paintings, the find of a snake tube supports the identification of the room as a hall used for ritual activities. South of A, a precinct (space B) was laid out, which functioned as an intermediary space between indoors and the court of the Agora, separating the house from the ‘public’ sector and possibly hosting at times groups larger than the capacity of space A.

12The sixth and last building phase encompasses the creation in LM IIIB of an elongated precinct to the south of P/Q, named W, and of adjoining spaces (E, F, and R), that do not communicate with the building (Privitera 2015a: 80). A major hindrance to make full sense of the function of W is the apparent lack of openings, which seems to turn it into a sort of hortus conclusus. At any rate, the presence of an entranceway cannot be ruled out, as a long stretch of the west wall of W does not stand to our day, having been razed when a Hellenistic structure was laid out. Indeed, this spot could represent the most suitable position for such an opening, which would have been located not far from that in the south wall of P/Q and next to a path leading westward and downhill through a small stair, three steps of which are still preserved to the south of space R.

13On the whole, a functional interpretation of the various spaces of Casa VAP is made difficult by the scanty floor assemblages related to the period of abandonment (Privitera 2015a: 74). The few artefacts excavated are mostly large storage vessels difficult to grasp and transport, that were probably left behind by the inhabitants at the time of the peaceful abandonment of the building. This is not contradicted by the discovery of a few ritual items, such as the snake tube in space A, which could have been originally embedded in the floor, like other specimens of the same class found at the site. Moreover, it cannot be excluded that space A, along with the nearby space D, could have been used or frequented even after the abandonment, as a large kylix excavated in the latter space, possibly LM IIIC early in date, seems to attest. Eventually, a few figurines from P/Q testify the carrying out of domestic rituals in this sector of the house, either on the ground floor or on the upper floor.

14In several regards, Casa VAP is quite a peculiar building. Firstly, this is the only construction of domestic character at the site, possibly together with a few more spaces that abut it on the north (Privitera 2015a: 85-95); therefore, it does not seem too farfetched to consider it as a sort of ‘Little Palace’, that is, the residence of the local ruling group, against the backdrop of an atypical settlement, which appears as a ‘Houseless Town’ (Privitera 2011). Secondly, its complicated building history, which encompasses the creation of stairs, inner partitions, and the blocking and opening of passages respectively, spans the entire LM IIIA2 phase, that is, a relatively short period of about 50 or 60 years. This seems to hint at the constantly changing character of the household/s that inhabited it. In particular, from the third phase onwards, Casa VAP was endowed with large halls (A and P/Q, possibly O as well), that originally commanded non-communicating sectors. It is not difficult to envisage that autonomous suites of rooms existed also on the upper floor, which functioned as ‘flats’ for the inner subdivisions of the local household or were functionally diversified (e.g. by gender or activity). Thirdly, the presence of two distinct sectors on the ground floor can be traced throughout the lifetime of the building, since the west sector comprising spaces G, L/M, N, and later O did not communicate with the rest of the building from as early as the second building phase. One can wonder whether these spaces had an administrative and economic function, or represented one more residential suite of rooms, initially centred upon the large space L/M and later upon space O.

2. Casa VAP as a Case-Study of Syntax Analysis

15Two characteristics of Casa VAP motivated the use of space syntax analyses in this paper: first, the whole building is fairly well preserved and presents a complex architectural history which has been recently documented by one of the authors (Privitera 2015a). Second, the building produced very little archaeological material, tying functional approaches of the structure almost exclusively to architectural considerations.

16In such a context, configurational analyses are particularly well suited. The specifics and theoretical basis of this particular methodology being fully developed elsewhere (Letesson 2009; Letesson 2013), sufficed to say here that space syntax – usually coupled with visual analyses – focuses on characterising the spatial arrangements between the constitutive parts of a built environment. In so doing, it aims at addressing the potential circulation and occupation patterns and, consequently, at shedding light on possible modes of encounter and avoidance between different categories of users of the building.

17Although space syntax methods are particularly relevant when applied to a wide range of buildings in a comparative perspective, in this particular case-study, it is the architectural development of Casa VAP which is the main focus of analysis. In other words, by tracing the evolution of the building in terms of spatial arrangement, we will offer interpretive remarks on the potential social relations of which it was the arena. The latter are meant to complement and/or nuance the functional hypotheses presented in the recent publication of the building (Privitera 2015a).

18Because of space limitation, the analytical material and its detailed interpretation are presented here in annexes. The text therefore only features a summary of the major configurational characteristics of each phase and highlights how they can provide insights on the possible functions of the building.

2.1. First Building Phase (Fig. 11.8)

19Such a small building does not lend itself easily to space syntax analysis. Nevertheless, some remarks are worth making. Configurationally speaking, if one considers that space L/M is, at the same time, of type d. very well integrated and visually controlling, one could argue that it was its pole of convergence, a space where users had equal access and could interact. In that case, spaces N and G/H/I could have been relatively independent rooms with their own entrance but an equal access to central space L/M, which might have been the centre of the building’s activities. As suggested in Privitera (2015a: 81), the fact that G/H/I featured ashlar masonry and had a centrally located doorway could make it the major entrance way into the house. In that case, space L/M seems the best candidate for the role of “a large hall commanding the inner circulation and located next to one of the major entranceways” (Privitera 2015a: 81), a situation which seemed to have been perpetuated during the second phase.

2.2. Second Building Phase (Fig. 11.9)

20Starting with the second phase, Casa VAP clearly became a more complex building. It featured at least two staircases and therefore, in all likelihood, an upper storey. The latter might not have covered the whole area (see Privitera 2015a: 78) and could have been associated with a roof terrace.

21The biggest change in the west part of the house pertains to the transformation of space G/H/I in three independent areas. Space G now is a dead-end space (type a) and is clearly subordinated to space L/M. Space H was probably framing the upper flight of the staircase which started in I, directly from the exterior. One striking feature of staircase I, echoed in staircase D, is the direct accessibility from the exterior. Staircases immediately opening on the exterior are relatively uncommon in Minoan architecture and this clearly contributes to make the upper floor relatively permeable and accessible. Space D is also a dead-end space.

22With the second phase, it therefore clearly appeared that room L/M was the central room on the ground floor (a role that it might already have had in the first phase). Nevertheless, with the introduction of a second floor, one has to consider that another important room (if not more) might have been located there as well. Furthermore, the introduction of spaces of type a (rooms G and D) is typically associated with occupation rather than movement. Such a property could indicate an attempt to seclude activities taking place in these spaces from unwanted interferences. Given the layout and the visual properties of the building and keeping that idea in mind, space G could certainly be interpreted as a storage area (for goods used in the household) or a sleeping area whereas space D would better be understood as a potential storage area (for goods used outside the building or in relation with activities taking place in the upper floor or outside). In such a perspective, given that no hearths have been excavated in the building (indeed, nowhere on the site), one can wonder whether the area to the east of space D, later to be occupied by space A, represented by this time a courtyard used for actions such as the processing and cooking of food, not unlike other contemporary buildings on Crete (e.g. the central building at Kephali Chondrou: Hayden 1990: 204-207).

23Overall, despite the fact that the building started to develop along more complicated lines, in this phase, it is highly integrated (value = 0,196), i.e. all the rooms are tightly connected to one another and the exterior.

2.3. Third Building Phase (Fig. 11.10)

24The third phase of Casa VAP is the first major extension of the building. It is characterised by the addition of two new large rooms (P and A) and a clear modification of the access system, with new entranceways and the blocking of earlier openings (the most important one being the south access to L/M).

25Qualitatively speaking, the graph expands in a tree-like fashion, with the single ring of the ground floor centred on room P, which clearly appears as the hub in the ground floor circulation patterns (see Privitera 2015a: 79), a characteristic which is also confirmed by quantitative and visual data. A room being dominant on a ring is usually interpreted as the interface between residents and visitors in a domestic structure. With its nice double entrance and its multiple accesses (both from outdoors and the upper floor of the building), room P is a very strong candidate for the central room of the ground floor. At this stage, it clearly became the centre of the building and, most certainly, the pole of convergence of the whole structure, the place where both visitors and residents met and where the different sub-groups within the resident category could congregate in activities that would have enacted group solidarity. In fact, with all the staircases leading to the upper floor, the overall spatial configuration is strongly reminiscent of the internal arrangement of many large Neopalatial houses (Letesson 2009; see also Privitera 2015a: 141). Judging by the overall layout and the potential extent of the first floor (Privitera 2015a: 79), it does not seem too far-fetched to posit that, apart from the N-L/M-G area, particular domestic activities might have taken place on the upper floor. Following Driessen (2005), one could also hypothesise that some ritual activities could have taken place there as well.

26Developed along linear sequences (N-L/M-G and A-D & staircase D), the rest of the building betrays a stronger tendency to organise space in a manner that facilitates the control of circulation (with type b spaces leading to dead-end spaces of type a). Of course, as suggested by Privitera (2015a: 79), the number of staircases converging towards the middle area of the upper storey indicates that a certain amount of ringiness might have also existed there. Nevertheless, he also observed that “it cannot be excluded that passages located on the upper floor were blocked in order to create independent units” (Privitera 2015a: 82). Furthermore, the high segregation value of certain spaces (particularly G, D and the sottoscala) probably hints at a tendency to seclude some spaces both from the rest of the building and the exterior. This is manifest in both the higher depth of some spaces but also in the overall integration value which is clearly lower than in the previous phase (although it remains pretty high for such a sizeable building). From the third phase onwards, the overall integration of the building (i.e. how closely connected to the exterior all the rooms of buildings are) will only decrease. This certainly testifies to some sort of functional change, especially by comparison to the first phase.

27Several complementary remarks can be made about Casa VAP as it developed into a much more elaborated structure. First, until the final abandonment of the building, its west sector (rooms N, L/M and G) remained an independent and isolated unit, only accessible from the exterior through room N. On the basis of some of the few finds identified in the area, Privitera (2015a: 82) suggested that these rooms might have been principally dedicated to storage. If this was the case, these storage areas were probably meant to provide supplies for activities taking place outside. Another option would be that this area – basically the ‘ancestral’ core of the building – was simply hosting one of the independent domestic units resulting from the “growth and inner subdivision of the local household” (Privitera 2015a: 81).

2.4. Fourth Building Phase (Fig. 11.11)

28With its fourth phase, Casa VAP gets closer to its most elaborate and penultimate phase. This phase is marked by two changes, a minor and a major one. The minor change pertains to the addition of space O (Privitera 2015a: 79). The major change is the construction of space C which, eventually, with the creation of the opening between C and D, made “the inner circulation at the ground floor [...] possible” (Privitera 2015a: 82). This phase is here presented in two plans because this small opening (Privitera 2015a: 43) might have been created after the construction of room C which, originally, might just have been a type a annex of room P. Although this opening was certainly too small to be a proper doorway (0.65 x 0.95 m), the fact that it was not designed originally but created afterwards is certainly significant. This opening was needed for some reason and although it was not probably meant for visitors, with phase 4 bis, it was possible, on the ground floor, to get from one part of the building to the other without getting out.

29Most of the remarks made about the third phase still apply to the fourth phase. Further observations are nonetheless necessary. Although the two fourth phases show almost the same graph, the opening between C and D transformed a tree-like structure into a ringy one. While the spatial configuration of phase four clearly testifies to the existence of three independent sectors on the ground floor of Casa VAP (west: N-L/M-G-O; centre: P-I-Q-corridor-sottoscala-C; east: A-D-staircase D), the creation of an opening between D and C clearly illustrates the desire to create strong spatial links between the east and central part of the building, which were previously separate entities that might only have connected through the upper floor. Although this could simply have practical causes (such as passing goods from one part of the house to another), this also attests to the creation of more complex circulatory patterns within the building. The latter, by allowing a larger variety of movements, both on the ground floor and from/to the upper floor, might indicate that, at that time, Casa VAP was more than a simple residence and might have hosted the activities of different groups (or sub-groups of an extended household), which could have then been kept relatively separate from one another. This type of configurational properties is also evocative of Neopalatial patterning of spaces and betrays an architectural elaboration that is not that common in Postpalatial buildings.

30The construction of space O in close vicinity to the independent unit N-L/M-G could perhaps be interpreted along different lines than those originally suggested by Privitera (2015a: 79), who interpreted it as a storage space, at least by the time of the abandonment of the building. If the linear sequence N-L/M-G truly is an independent domestic unit – with its own storage areas – within Casa VAP (rather than a pure storage area), then room O might be interpreted as a social space where the inhabitants of the west sector of the building might have entertained their guests or gathered at specific occasions, maybe on the low benches existing in the room. In fact, whereas low benches are features of storage areas, the presence of the round stone base of a wooden pillar in this room (Privitera 2015a: 52 and 139) seems to be reminiscent of the only other specimen found in the building, that in space A. Accordingly, one is left to wonder whether space O originally shared the same receptive character as the former, prior to the substantial refurbishment that A underwent in the fifth building phase.

2.5. Fifth Building Phase (Fig. 11.12)

31With its fifth phase, Casa VAP really reached its acme. At that time, the building took its most elaborate form and, as noted by Privitera (2015a: 80), it was truly integrated in a wider urban project which also comprised the Agora and its propylon. In terms of layout, the major change is related to the construction of the courtyard B which was added to the south-east corner of the building. Another noticeable addition is the construction of a long wall bordering a street leading to the twofold entranceway opening onto space P, which clearly was the main entrance to the building (Privitera 2015a: 80).

32Qualitative properties of the fifth phase are closely similar to those of the fourth phase and it is only the introduction of room B which made a small difference. The construction of the courtyard indeed contributed to extend the external ring and, consequently, established a higher potential of access control towards space A, which was then considerably monumentalised (its ceiling was raised and its walls adorned with wall paintings). Apart from that, the overall configuration is just the completion of the architectural arrangement which first appeared with the opening between rooms C and D during the fourth phase. Of course, as suggested above, it is worth mentioning that the particular character of this opening hints at other types of spatial behaviours than a normal doorway. Although taking it into account does not really affect quantitative and visual properties, there is indeed a stricter built-in spatial control in a tree-like structure than in a ringy one. One way to interpret this particular opening would be to suggest that it was not meant for ‘visitors’ (defined in space syntax as people having a temporary access to the building, under the control of the residents and a social identity which manifests itself collectively), but rather accommodated activities of the ‘residents’ (defined as people whose social identity is integrated in the layout of the building over which they have a certain degree of control and to which they have a privileged access). Indeed, one is left to wonder whether the creation of such an opening echoes the substantial transformation of space A, the new ritual functions of which could have substantially affected the circulation patterns on the ground floor – all the more so if the upper floor was subdivided in non-communicating sectors.

33In terms of visual analyses, as staircase I is directly visible from the double doorway into room P, one could argue that it might have been the more ‘public’ access to the upper floor. But, if one considers that staircase Q was also fairly visible from this entrance, this creates an unusual situation where two accesses to the upper storey are concentrated in one single room which is also the main entrance of the building. If the overall configuration of room P (with its north-east entrance, staircase Q and its southern doorway) is evocative of the central room of the Neopalatial North House at Kommos (Letesson 2009: 182-186), the existence of staircase I makes it a very uncommon type of room (reinforcing its particular character and, probably, the essential role within Casa VAP).

34During its fifth phase, Casa VAP certainly was at the peak of its architectural development. It formed a complex architectural structure with differentiated circulation routes, both on the ground floor and towards the upper floor, that probably attest to different categories of users and/or categories within the extended household group. These circulation routes are also characterised by visually controlling points, especially at the connection between rooms, that could be evocative of a need to channel or monitor circulations throughout the building. Furthermore, the building was clearly monumentalised, in that it was integrated in a wider urban project and presented an accessibility more conspicuously channelled than before (most notably with the street bordered by a long wall leading to the main entrance in space P and the courtyard B which formed an interface, a buffer zone, between the Agora and space A). When compared to the other LM III A-B buildings known to date, the fifth phase of Casa VAP really stands out as an achievement both in terms of architectural design (spatial configuration) and building techniques.

2.6. Sixth Building Phase (Fig. 11.13)

35The last phase of Casa VAP represents by far its most radical transformation. Indeed, by closing the south access to the house through room P and creating a large internal courtyard (precinct W), the whole configuration was profoundly affected. Of course, as suggested by Privitera (2015a: 55), an opening might have existed in the west wall of the precinct. Such a doorway would create a general spatial configuration that is not truly different from that of phase 5, the whole arrangement being only modified by the appearance of an additional space of type b/c before the main room P. As this opening is only hypothetical – although fairly probable – both reconstructions are briefly analysed here.

36If the west opening into precinct W did not exist, with the sixth phase, the complex configuration of Casa VAP unravelled into linear sequences which betray totally different types of interfaces (spatial relationships between residents and visitors) and most certainly household dynamics than in previous phases. This phase is evocative of a less subtle way of channelling/controlling/managing the circulations within the buildings. This could mean that, before its abandonment, Casa VAP might have retrograded to a mere domestic building with very few interactions between residents and visitors and among the different components of its household. Alternatively, it could also suggest that only the east part kept its ceremonial vocation whereas the rest of the building became purely domestic and/or was progressively abandoned (as is the case at Sissi where, during the last LM IIIB phase, the ritual centre of the building was one of the few areas that was still in use). In that case, as noted by Privitera (2015a: 55), the precinct “might have functioned as a sort of hortus conclusus of Casa VAP”. On the contrary, if this opening existed, this sixth phase was not that different from the fifth one and all the remarks made above concerning the latter would apply to the former. The only difference would then pertain to the creation of the precinct itself which would have simultaneously blocked the street and the direct entrance to room P, hence lowering the overall monumentality – or urban scenography – of the whole area. Nevertheless, being particularly well integrated, this space could also have hosted particular types of social gatherings and/or played in relation to room P a role similar to that of courtyard B in relation to room A.

Final Remarks

37Thanks to both its central location within the settlement and the monumental layout it had at the end of LM IIIA2, Casa VAP no doubts stands out against the contemporary Cretan domestic built environment. Despite the lack of floor assemblages, caused by both its abandonment and the excavation techniques in vogue during the early 20^th century, its complicated building history makes it a real conceptual ‘labyrinth’. Accordingly, space syntax analysis represents a useful heuristic tool to make enquiries into the various ways human engagement could have affected such a building. In fact, given its safe centrality within the site, the focus of our enquiry was to enrich its interpretation by focusing on notions such as accessibility, movement and visibility, as clues of how human behaviours may have transformed the built environment across time and space.

38On the one hand, formal analysis makes it clear how much the ‘genotype’ of the house owed to the Neopalatial building tradition – indeed, more than one could expect in LM IIIA2 Crete. In such a perspective, it is worth observing that Neopalatial features do not seem to appear in the first two building phases, but only start with the third one: the presence of many stairs leading to the upper floor, the overall configuration of space P, the general circulatory pattern during the third and fourth phases. If Casa VAP was originally built to host members of the Mycenaeanised Knossian bureaucracy (Privitera 2011), it is all the more important to emphasise the increased impact of traditional, ‘indigenous’ behaviour across the building stages which could reflect the diminishing control of Knossos on the Messara plain and eventually its eclipse by mature LM IIIA2 (Privitera 2015a: 146).

39On the whole, the graphs of the first two building phases clearly show that the building initially developed according to an agglutinative mode: Casa VAP was easily accessed from outdoors, and it remained as such when an upper floor was added, as the presence of stairs directly opening onto the exterior clearly shows. From the third building phase onwards, the creation of new large halls located next to the major entrances represented a clear turning point, marking the passage to a proper articulated mode. As in several large Neopalatial mansions and villas, the creation of staircases determined the formation of diversified ‘paths’ inside the building, which connected limited sectors of both the ground and upper floor by drawing on a stronger demarcation between indoors and outdoors. Unlike Neopalatial architecture, at any rate, the concern about possible intrusions did not take the form of porticoes or vestibules, but that of two courtyards (B and later W), which both have sensibly framed the relationship between the large halls (P and A) and the nearby court and street respectively. The tendency towards a progressive integration between spaces on the ground floor, at any rate, was made difficult by both the complicated ‘growing up’ of the building and by its location across a slope. From the fourth phase onwards, the ground floor was clearly divided into three major sectors centred upon two, and possibly three, large halls (A, P and perhaps O). The need to connect the east spaces (A, B, and D) and those located to the south (C, P, and Q) safely inspired the opening of the weird passage between C and D. One can accordingly wonder whether such a passage was created prior to the refurbishing of space A (phase four-bis, suggested above) or was part of a single, wider rearrangement of the house (fifth phase). At any rate, it seems safe to conclude that such a passage, if hidden and inelegant, made it eventually possible to the inhabitants of the house to escape their own labyrinth or, to put it simply, to get out of their dead-end spaces.