Locations in the Vostitsa Landscape

In this episode, we explore differences in the landscape, specifically around locations in the three datasets (1463, 1700, and 1907) by comparing them with each other and with a dataset of 50 random points. The analyses show trends in the landscape of locations through time.

The maps and plots are available on the Helonaki website.

Transcript

Jen: 0:00

Before we arrive in Vostitsa, we traverse a high mountain and then proceed for a league along the fine valley, which extends a long way to the south. This is the way to go from Vostitsa to Dimizana and into Upper Arcadia. There are many villages all under the jurisdiction of the Pasha of the province and governed by sub-bachis or syndics of his appointment. The strata of the mountains incline toward the north, as if the earth had sunk on this side to form the Gulf of Corinth. That's Pouqueville writing of his journey in 1799 to Vostitsa in his 1805 book Travels in the Morea, Albania, and other parts of the Ottoman empire. Which was published in his native French. This is from page 75 of the 1813 English translation by Anne Plumptre. I'm Jen Glaubius, and this is the Helonaki Deep Dive, a podcast about mapping and analysis for historical and archeological research. In this episode, I discuss settled places, mostly villages during the three time periods we're looking at, by putting those locations into context in the landscape. So looking at what trends we can see in the landscape around those locations. And this of course is building off of the last episode which looked at the environment of Vostitsa as a whole. And I just wanted to mention the plots generated in the analysis can be viewed on the Helonaki website. A link to the page with the plots and maps is in the show notes. Let's dive in. Before we get into the details about the Vostitsa landscape and environment itself, I first want to talk about the relationship between humans and the environment. We know that humans impact the environment in many ways, especially. Today we build dams. We divert rivers. We build terraces. We build all sorts of constructions. We farm the earth. We impact the environment. And of course the environment impacts us. Different natural processes, hurricanes, floods, landslides, all of those things impact human activities and where humans will do things in the environment. But I want to talk about a geographic theory that was very influential in the 19th and early 20th century, which was called environmental determinism. This theory was that the environment. Especially climate in some cases, determined human culture and activities. That humans did things because of the climate, especially, or because of their environment. And this originated from ancient Greek philosophers who thought that climate zones influenced how humans act. Which was racist. By saying that peoples in warmer climates, were lazy. It was picked up by modern philosophers and geographers, especially in Northern Europe. And used to justify racist beliefs about the Mediterranean and African cultures, especially. So thankfully environmental determinism, this theory is no longer a mainstream theory. But the ideas from it pop up among racists. So if you end up seeing any discourse that says, oh, they're from this very hot climate, they must be lazy it's entirely this racist, environmental determinism. But that's not to say that the environment doesn't influence human activities, it does. But the environment does not determine what humans do. Because humans have agency, we make decisions about our own actions. There's no predetermination based on the place, the landscape, the climate, the environment. Instead the environment provides possibilities for humans to act upon. One way to think about it is as if the environment was your pantry full of ingredients for baking. So based on where you live. The ingredients available in your pantry might differ. So. So you might have some more fruits or some other things, but even if you have the same staples of flour, eggs, butter, milk, and yeast. One person might make bread while another person would bake cake or cookies based on their own skills and what they have decided. And so humans will do different actions in different environments based on, are there trees here? How about available stone for building? Where are the streams? Are there not many streams and different things like that? What is the climate? What can I grow in this climate? Other things like that. But humans in the same environments do not always act in the same way. Because human decision-making human agency and societal pressures, also play a part in the human use of landscapes. When we're looking at a landscape, the ways that people use it will differ because it's based on their needs or on other things like that. NOw In the last episode I described the landscape of Vostitsa, particularly the terrain using GIS analysis of a digital elevation model or a DEM. In this episode, many of the types of analysis are the same. So I'm using all the analysis I had generated. But now, instead of looking at the area as a whole, I'm going to focus on the areas around the locations found in the 1463, 1700 and 1911 data sets. And I'll also be adding in a couple more analyses. So overall, what this episode is about is looking at the use of those places. So use of the landscape and how settlement patterns might've changed through time. So to start off with, let's just talk about a few of the theories about settlement patterns in early modern Greece. An early theory was the idea of height zone, nation, or demographic retreats. Which was during unsettled times, especially during the Ottoman period, Greeks retreated into the mountains. And so there were more settlements at higher elevations, which is the height zonation, that there were zones where people settled because they were in the mountains and thus safer from the Ottomans or from pirates. But later, towards the end of the Ottoman period, or definitely after the Greek war of independence that then villages were moved more downslope and closer to the coast. There was an analysis of villages in early modern Greece by Wagstaff in 1978. We can also compare that with, a study by Forbes in 2007. Looking at the peninsula of Methana, which is on the Eastern side of the Peloponnese. Where there isn't necessarily movement to higher elevations. But, more to hidden places, places that are harder to view from the coast. So there is fear and insecurity that's driving where settlements were located, but it doesn't necessarily correlate with elevation because there's other things about the landscape. It's not just elevation, but perhaps how hidden a place could be. And I'm pulling, especially from a study by Seifried in 2015. Who looked at settlement in the Mani, which is on the Southern part. And Seifried looked at elevation, but also looked at other factors. Like How connected were sites? How many villages were within three kilometers radius of other villages? So within, for each village, how many other settlements were within three kilometers. Seifried also used distance to the nearest accessible bay. The Mani is a peninsula. They have many bays. As you'll hear about in just a bit. I instead use distance to the coast since Vostitsa is not a peninsula. And the coastline makes more sense than distance to the bays. And Seifried also looked at number of inner, visible settlements. So how many settlements could you actually see? And I frankly did not have time to run that for every single one, because you have to do it manually. So I'm going to be looking at many of the same things as Seifried, because I'm interested in changes in the landscape and use of the landscape through time. Let's talk a little bit about the analysis. We have three data sets. There's the 1463 locations in the Ottoman defter of Vostitsa. Of those there were 55 known locations, locations that I could actually plot on a map. But only 23 of those were within the 1700 boundaries of Vostitsa so a little under half. And more were within Kalavryta. But there was one settlement that belonged to Kalavryta, that actually is within the 1700 boundaries. So there are 24 locations in the analysis of 1463 This is the smallest of the three data sets within Vostitsa. For 1700 because we're using the 1700 boundaries, there are 57 total locations that I've gotten from the Venetian map. Also dating to 1700 of the territory of Vostitsa. There's 57, but two are excluded from the analysis because they're named paleo Castro, which is old castle. And so those were no longer in use during 1700. One of those locations seems to match up with a 1463 settlement. Which is interesting. And the other one doesn't and I'll probably revisit that when I talk about population in a few episodes. All right. So 24 for 1463, there's 55 locations for 1700. In the 190 7, 1911 census records, there are smaller areas that are all within the province of Achaia. There are locations within the 1700 boundary from Aigio, Akrata, Voures, and Krathidos. So there's 63 total within the data set, but four of those are not within the 1700 boundaries. Some are off on the west side, a few on the east. And so in this analysis, I'm using 59. So this is, this is the largest of the three, but only four more than 1700. In addition to all of those locations from the three data sets. I also generated 50 random points in the landscape, in places where there were not settlements. So deliberately not close to any settlements. I did this because I wanted to see if there are differences in factors from the random locations where there weren't settlements to the settled places that we know about. Those 50 random points will be used in this analysis and you'll see it in the plots, but not for all of them, because it doesn't make sense for certain things. And lastly, when I was examining most factors, I looked at a mean value within one kilometer of the location. So not just the point on the map, but actually within one kilometer of that, because that better includes the actual settlement because the settlement, isn't just one teeny tiny little point, but an area. And so one kilometer is probably a little bit large, but it gives us a better general idea of what was going on in a settlement than just trying to look at what was going on in a small little point. And once again, you can see all the maps and plots generated for this analysis on the Helonaki website. And that link is in the show notes. So go look because the plots are kind of cool. All right. So the first thing, I looked at was the height zonation, is there a change in elevation and mean elevation through time? I looked at mean elevation within one kilometer. And examined it using box plots. If you look at the box plot, you see that the 1463, 1700, 1907 locations have the same overall range. So the lowest and the highest locations are in the three data sets. Although the quartile range, which is the box itself, and the median, which is the line in the box, all of those are different. The random locations have a larger range and median close to the 1463 median. You see a change through time. Median elevation gets lower. It's highest in 1463 at 500 meters above sea level. It's down to 400 meters above sea level at 1700. And then the lowest, just under 300 meters above sea level in 1907. Median elevation, the midline in elevations, goes down through time. The quartile range, which is the mean elevation for the majority of settlements, it's smallest in 1463. That's the smallest range, they're all kind of packed closely together between 300 and 600 meters above sea level. The largest quartile range, the box, is for 1700, between 150 and 700 meters above sea level. So you see some settlement lower down, but also slightly higher up than in 1463. And then 1907 is slightly smaller in range, but it's also shifted downwards to less than 100 to 550. So it seems like after 1463, there were settlements at both higher and lower elevations, but really in 1700. And then especially by 1907, the trend is for settlements in lower elevations and some of the higher elevation locations were probably abandoned. So that's elevation. Let's think about ruggedness. So this is how much up and down did the landscape do. The random set of locations has the greatest range of ruggedness, including outliers. The median ruggedness didn't really change much between the 1463 1700 and 1907 sites, except it's just a little bit lower in 1907. Slight trend for less rugged locations later on. The quartile range expands towards less rugged places in 1700 and 1907. It's pretty compact in 1463. And then it becomes less rugged through time. So choosing locations that are a little more accessible, lower elevations and not quite as rugged through time. All right now let's think about landforms. Landforms have to do with the type of landscape. Is it a fairly flat, plain? Is it on a sloping area? Is it a valley, is at the top of a ridge, which would be very visible. Areas that are less visible, maybe along a slope, especially a shoulder of a slope, but it's harder to build on. Things like that. And the streams, which would probably be liable to flood. If you look at this plot, which is not a box plot, but a bar chart. You see that there are fewer locations next to streams through time. And there are none by 1907. There weren't many to begin with, but less building near streams and none at all by 1907. So using different locations. And new locations that you get through time after 1463, you get them on slopes, in valleys, which would be fairly hidden, on the plains, which would be more visible. And then on ridges, which would probably be more visible. But it would also be a little difficult to reach, to actually access that area. The increase in locations in valleys and plains, which are probably fairly accessible, could point towards a decreasing need for security. So as time went on, there was less need to have your village in a secure location. So those are the location for some of the environment

17:34

factors.

Jen: 17:36

Let's look at a few other factors. Neighboring settlements. So how many other settlements are within a three kilometer radius, which is a little less than two miles And for this one, the random data set is excluded because this has to do entirely with human decision-making. The other ones do too, but it didn't make sense to plot out the random data set. From the plot, you can see that the least interconnected period. So the period with the least settlements with neighboring settlements. Is in 1463. If they have any villages within three kilometers, they only have one to two and one outlier is within four other villages. But that's unusual. The 1700 locations are the most interconnected. They have a range going between one and 10 other settlements within three kilometers with a median of four. So. The median value for number of settlements within three kilometers is four. That's a lot of neighbors for many of these places. In 1907, it goes between the other two. So it's more interconnected than 1463, but less than 1700. The range goes between one and seven villages are within three kilometers with a median of three villages within three kilometers So for this one, there's no temporal trends. It doesn't change in one direction. It goes from least in 1463 to the most in 1700 to in between the two in 1907. Let's look at one more analysis. Which is the distance to the coast. So how close were villages and the town of Aigio to the coast, which Aigio is on the coast. So it's the closest. Looking at these box plots. The random data set had the greatest range of distance to the coast. They had one that was 20 kilometers from the coast. So almost at the very edge of the 1700 boundary. And it had just the greatest range. Otherwise, there's a trend towards locations moving closer to the coast by 1907. There's more settlement closer to the Gulf of Corinth by 1907. In 1463 and 1700, they have almost the same distance. Except the median in 1700 is slightly less than 1463. So. There's movement closer, maybe a little. Lower elevation but overall by 1907, people were living closer to the coast. In more accessible areas. So, what can we take from this? From all of these analyses that I've described and that you can go look at on the Helonaki website. There is a trend towards lower elevation through time. The settlements come closer to the coast and they're on the plains. More of them on the coastal plain by 1907. That's not the case in earlier time periods, there was a less settlement on the coastal plain and close to the coast. And that was probably because it was too accessible, even though it's a nice place for agriculture, it's also really easy for people to get there So settlements in the two earlier time periods, 1463 and 1700 were less accessible. This is very evident. I drove around the Vostitsa area in June of 2022. I was trying to match up places from especially 1700. And it was harder to reach the areas further south. Further from the coast, more into the mountains. And many of those are no longer settled. So there was no place for me to actually get to, or it's hard to know if I was there. The number of neighboring settlements within three kilometers of each other, increased between 1463 and 1700. This was due to an increase in the number of settlements overall. So you're more likely to be closer to other settlements when there are more. The data set for 1463 is the smallest, it's half the other two. So it makes sense that there aren't as many settlements close to each other because there aren't as many settlements overall. That also means that within this area, there just weren't as many settlements and they weren't all in one location, they were more scattered. Using different parts of the area. This gives us a general picture of changes in where people settled through time. But keep in mind, this is just locations. And there's another aspect to this that we'll get to in a few episodes and that's population. Just because there's a settlement, it could be like five people versus a settlement that had 200, 300. So you can't entirely see people with settlement locations. You need to actually look at population to get the full picture of what was happening with settlement. If it was actually more people up in the more remote distant areas or down by the coast. So there's another aspect that needs to be added in which is population. But we'll talk about that in a few episodes. So stay tuned in. In the next episode, we'll talk about the town of Vostitsa today known as Aigio and how the town changed through time. And now for some endnotes. I looked at a few different papers that looked at settlement patterns in early modern Greece. There's been a lot of work looking at very ancient patterns in the archeological record. But because we're dealing with the early modern, I especially use the papers by Wagstaff the 1978 height zonation paper. Forbes 2007, who had a critique of Wagstaff's analysis, and then Seifried 2015, who brings things together and I've followed some of Seifried's analyses. Thanks for listening. Email questions or comments to deepdive@helonaki.com or ask them on the Helonaki Deep Dive Facebook page. Show notes with links to resources mentioned in this episode will be available at helonaki.com. That's H E L O N A K i.com. You can also find ways to support the show now, including merch, such as t-shirts, mugs, and stickers with the Helonaki Deep Dive logo at helonaki.com/support. My thanks to Patreon supporters at the geospatial analyst level, Leah Varrell and Janice and Jerry Farrell. Your support keeps the Helonaki deep dive going. The Helonaki Deep Dive is written and produced by me, Jen Glaubius of the Helonaki. The theme music is Deep Ocean, instrumental by Dan o of dan o songs.com, additional sounds from zap splat.com. Thanks for listening.