What are the significant ways in which Classical Greek and Roman houses differ? What can we learn about their households from these differences?
It is important to remember that houses in the Classical and Roman periods need to be analysed carefully. Excavations carried out through the ages have had varying degrees of accuracy when interpreting the information that is gleamed from the artefacts. Allison describes how items are ‘decontextualized' and says that very often sites have been removed from their situations before the context has been properly recorded (2004: 4). This needs to be taken into account when we compare houses and their included artefacts, and also how we interpret what these things tell us about the contemporary societies. While on the subject it is important to account for the fact that some items of less artistic merit may have been removed for museum collections and those that are even less interesting have been left in situ, this can provide an unrealistic interpretation of the site.
While investigating the different types of housing I will be using several case studies, namely for the Roman topic I will be using Pompeii for examples. It is also important to remember that we have a lot more standing evidence for Roman housing than we have for the Greek housing. In evidence for Greek housing we have mainly floor plans whereas we have preserved sites such as Pompeii and Ostia (Italy) for the Roman contexts. This essay will not cover the remains of the houses but rather what the houses would have been like, and will provide a comparison between the standing structures, and is not intended to discuss the differences in preservation of the sites.
Evidence we have for classical Greek housing is very limited. The structures do not survive like the examples we have of Roman houses that still exist. Although Pompeii is unusual it means we gain an undisturbed look at a society, and can explain a lot about the way of life. Unfortunately we do not have many well preserved sites for classical Greek housing so we have to get the information by thorough excavation of sites and extracting the information from the artefacts found. One example that is often quoted by historians in studies is Olynthus (Ault et al., 1999: 46 & Adrianou, 2009: 5). This will be covered in this essay along with buildings from Athens (Goldberg, 1999).
As there is less information for Greek housing than there is for Roman, making judgements on what the households were like by studying the housing is difficult. This essay will address housing from the two periods from the urban settlements of both cultures. The focus will be on urban settlements because there has been more research into these areas, therefore patterns can be more easily identified and more accurate conclusions may be drawn.
Evidence for housing within the Classical and Roman periods is limited as mentioned above, however, this does not mean that studies have not been carried out, and interpretations of the evidence uncovered have been made. In Athens ‘a small number of houses that are badly preserved' (Goldberg, 1999: 142) have been uncovered and the layout of several buildings have been surmised, especially those nearer to the agora (Goldberg, 1999: 142). In parts of Europe which would once have belonged to the Roman Empire there are examples of houses which are better preserved, for instance, Pompeii, which is a beautiful example, although not necessarily a ‘typical' Roman city. This essay is to focus on these areas because they are areas that have been most researched and the discoveries made here have been analysed most thoroughly with reports being published on the findings.
In Roman housing, when interpreting evidence of the use of rooms it is brought to our attention by Allison (2004) that rooms were labelled as they were excavated in Pompeii. This means that the name that has been allocated to these rooms is not necessarily the nomenclature given by contemporaries of the society. This has other repercussions, it means we cannot assume that the room was used for the reason that we assume from the name. For instance just because we label a room as being a dining room does not mean that it was necessarily used only for dining in. Sometimes rooms had a number of functions (Allison, 2004: 63).
There is more evidence of some houses than others in Roman societies. For instance, atrium houses are much better documented than others. McKay also describes the Etruscans as having atrium housing. This shows some overlap between the Roman and Greek societies. But he attributes these techniques to the Near East (McKay, 1998:15 & Palagia, 1998: 40). This is down to the fact that they are generally bigger than other houses. Due to this, it is true to say that in Pompeii they were excavated more carefully because they were distinctly larger (Allison, 2004: 29). Vitruvius describes how there were three types of atrium courtyards. One had a ratio of 3:5, another had one 2:3 and the other was 1:1 (VI.III.3). This is important because it shows there were rules to be followed when it came to Roman housing.
Vitruvius (On Architecture) describes five different types of courtyard but also describes them and the precise measurements of the rooms adjoining these separate courtyards. This would suggest that these houses followed patterns when they were built; and that there was relatively little difference between them.
Roman houses were built following more rigid guidelines than those expressed by Greek house plans. It has been described that Greek housing followed no pattern (Cahill, 2002), this is especially true at Olynthus where Cahill carried out a study on floor levels (2002), but is also seen in plans from houses in Athens. Figure 1 shows one of the houses from the north side of the Areopagus in Athens and is typical of the houses that have been exposed (Goldberg, 1999: 144). As shown, there is no central room which all of the others lead off from, unlike in Roman houses which have a central atrium which the other rooms open out onto such as the House of the Faun shown in Figure 2.
Roman houses seem to follow more of a plan demonstrated in Virtuvius's On Architecture which describes the definitive ratios that rooms and courtyards have to be.
Although the layouts are very different they have the similarity of both having a courtyard situated in the house. The function of this room in both societies varied, taking into account the time of day, year and what was going on in the house at the time. This is important as it refers back to Allison's point (2004: 63) which said that rooms often had a number of functions which could change. Goldberg also makes the point that this is the reason for moveable furniture (1999: 157); it makes it much easier to change the function of the room.
Roman houses have a number of rooms which tend to be found in most other houses. For instance, they all have atrium which lead back into other rooms and generally speaking have a peristyle behind this. The atrium is where most of the business side of things would have been done. This means that private and public matters were kept separated. Alternatively in Greek houses men tended to have a room set aside but this was not necessarily at the front of the house and meant that business was dealt with at home some of the time (Goldberg, 1999: 142). Goldberg (1999: 155-156) also states that the courtyard of the house was the hub of activity. Everything passed through here; even though it may have been a female domain it was a way for husbands to keep an eye on their wives and all traffic through the house would have passed through here initially. This is important because it demonstrates mistrust in women, who were thought by some to be difficult to predict and generally difficult (King, 2005: 110).
One of the main differences to be identified between the two types of housing to be studied is that Greek housing seems to have rooms that are gender assigned. This is not a theme that has been identified in Roman housing. It has been suggested by historians (2005: 231) that Greek women were more suppressed than Roman women, which Goldberg (1999: 158) argues may not have been true with the counter argument that women had some leverage and power over their husbands because of the dowries paid to their husbands at the time of their marriages (Goldberg, 1999: 158). What is not argued is that women had less power than men, was it then for this reason that men had different rooms from women? For instance, the andron was a room that historians have related to the male domain. It is thought that this was where symposia would have taken place (Goldberg, 1999: 149); however, this is not a phenomenon that is mentioned in Roman houses as having been something that was prevalent. Maybe this is due to the idea that women in Roman society were deemed to have had slightly higher standing than their counterparts in Classical Greece (2005: 231).
Ault et al. complain in an article published in 1999 that ‘both artefacts and architecture are studied as isolated entities' (Ault et al., 1999: 45), but what there has not been enough investigation into is the link between the two and what each one can tell us about the household as an entire unit.
As alluded to above, the open areas within houses in both societies being looked at were busy places within the house. Within the Greek houses they were a ‘defining feature of Athenian houses' (Jameson, 1990: 179) but also served a wider purpose, as a temperature control for the rest of the house and were a tool within the economic goings on of the society in that they ‘served as enclosed yards to ensure the protection of the household property' (Goldberg, 1999: 144).
The courtyards were considered to be the women's domain, although it was not unusual for men to be found here and it would have been used as a thoroughfare for male visitors wishing to get through to the andron (Goldberg, 1999: 147). It is only in more recent years that fewer assumptions have been made as to the value of each of the rooms. By looking at the evidence again historians are now better educated as to the functions of items of furniture and where they fit within rooms and what this tells us about the households and to a certain extent society.
These gender divisions which have been described by Goldberg (1999) are not always as clear as it would appear, for instance, the spaces within houses which have been assigned to females are not actually marked archaeologically by ‘women's objects, like mirrors or jewellery boxes' (Goldberg, 1999: 149). However this works in the opposite direction when assumptions have been made inaccurately about the function of certain houses just because there is a presence of one particular artefact, for instance loom weights. This is a topic also covered by Allison (2004) within the context of Pompeii, where inaccuracies were made about room nomenclature. The presence of loom weights does not necessarily presuppose that the house is a weaving shop (Allison, 1999: 70); it could be that there were just a larger number of looms within a particular house and the theory that they were mobile would mean that they could quite possibly have stored more than one or two looms in one household (Allison, 1999: 70). What it is more possible to assume is that the presence of loom weights in certain areas of the house, such as the courtyard, would suggest that these areas were dedicated to females (Allison, 1999: 71). In Roman society women would have done the weaving in the forecourts of the house as this was the ‘well-lit part of the house' (Allison: 1999: 70).
In comparing the houses from the two societies being studied it is clear that there are some spaces that one society demonstrates that the other doesn't. For instance, in Greek houses wells for water are frequent (Goldberg, 1999: 153); this is not something that is mentioned within sources on Roman housing. Neither did Roman houses include a room just for the purpose of male entertaining. Even the atriumwasusedby women for weaving (Allison, 1999: 71).
It is perhaps also worth noting that from the sources included within this study there has been no mention of urban villas having a second floor. However, there are examples where houses are situated above shops such as in Ostia (Storey, 2001) and are raised off the ground. This is also difficult to verify within excavation reports purely because if the building no longer exists then there may be evidence of a floor plan for the ground floor, but no evidence of the second floor would remain.
With studies like this one we encounter problems. To really investigate this topic, more research needs to be done which links the artefacts which are uncovered and what this can tell us about the household that they were found within. It is not safe to assume that just because an item was found in a room that this is where it belonged long term, an excavation is merely a ‘diachronic sample of debris reflecting patterns of use and behaviour over an extended period' (Ault et al., 1999: 52), and this snapshot of the household may not be entirely accurate.
Through the course of writing this essay it has been observed that conclusions are difficult to draw due to the nature of the material being dealt with. For instance, the irregular layout of Greek housing means that patterns are not easily identified as they are in Roman housing, there are of course similarities between them and patterns in the rooms that most often appear but there is no rigid layout which means we can predict what we will find, for instance, not all houses had andronesand some houses had second floors whereas others did not. Another fact to be taken into account is that a lot of the uses of these rooms is speculative. There is little evidence from primary sources from the time about the uses for rooms, so where historians have suggested a use for a room they are doing so by using the artefacts which is not always accurate (Allison, 2004).
It is difficult to directly compare the two types of housing as the Greeks and Romans go about their housing in different ways, with the Greeks dividing the house into genders, something which does not happen in Roman architecture.
This is a very limited cross section across the two societies and their houses leading to the conclusions being limited to urban houses and poorer houses may have been different again. This would be something to look into further. Therefore, ‘we remain woefully uninformed about many of the patterns of social and economic relationships within and between households' (Ault et al., 1999: 44).