The Assouan Dam

The topic I chose was the significance of the Assouan Dam in the Egyptian Gazette. I chose this topic because I thought it would be interesting to research something that is not inside Cairo or Alexandria, which makes up a majority of the Gazette"s news. When looking for a topic, I found trouble trying to find something that was significant on a global scale or at least outside of Egypt and the Gazette. I found the Assouan Dam when I saw almost an entire half-page dedicated to it, and after further research found a story that had continued throughout the four years of news that is available. Assouan was an area that was both rich in agriculture as well as tourism and historical importance.

Located on the border of the Upper Nile and Lower Nile, a dam was built in 1902 to help the agricultural industry as well as to protect Philae, an island in the Nile with an ancient temple complex. The temples were built around 6th-7th Century BC and is Assouan"s biggest tourist attraction even to this day. Due to Philae being a small island that had little elevation, it was prone to flooding if there was no control on the water level. Cotton, grains, and lentils were major crops that would be exported to Cairo and presumably around the world. It required the Nile to be at a steady level so that the fields would not get flooded. In early 1905, articles in the Gazette showed people complaining that the dam was not high enough and floods would eventually come and ruin Philae and the crops. However, the government did not care, and they did not raise the dam. In 1906, architects or scientists would show how the flooding is on track to breach the dam. In late 1906, letters to the editor included some stern messages to the government demanding for money to be allocated to the raising of the dam, still with no success. In July 1907, the dam flooded and ruined a lot of fields and flooded Philae, causing extensive damage.

This was included in the Gazette likely because Assouan was a major agricultural hub that would supply Alexandria and Cairo with crops. When I researched other reasons Assouan was included in the newspaper, the most common appearances were in the export and import sections, as well as stating when ferries or boats were departing from a port to go to Assouan. This was also included because the federal government was responsible for allocating funds to the construction and re-raising of the dam, not a local government. The “Letter to the Editors” section included several messages from people who were complaining or demanding these changes in Assouan, so that made it a big deal up in North Egypt where the Gazette was published.

This analysis project looks at the Egyptian Gazette to see what context the Assouan Dam is in throughout the newspaper. Originally using the “Find/Replace” tool inside Oxygen XML, I came across a lot of improper results for a lot of key words or phrases. Another issue I came across was word capitalization. I added a tag in my queries to make the word searches not case sensitive. Lastly, another tweak I made when querying was for the word “dam” adding spaces before and after ( dam ) so that it would pick up whole words rather than parts of words such as “Damascus”. Instead of finding how many times each word was said as well, I decided to count instances based on the number of sections included each term. This created a constant for all words as well, just showing how relevant they were instead of how common a word was stated. One section could repeat “Assouan Dam” ten times and be as relevant as a section whose header is “Assouan Dam” and it not stated again in the paragraphs.

The XML query format I used throughout this entire project was //div[@type="page"]/div/p[matches(., "assouan", 'i')] or if there was multiple words I needed to search for, //div[@type="page"]/div/p[matches(., 'assouan', 'i') and matches(., ' dam ', 'i')].

Prior to querying, I read about 20 sections that talked about the dam in order to see what was relevant to the topic. I found that a majority of sections about the dam talked or referred to one of three topics: agriculture, architecture, or the government. After initial querying, I found that some sections would not show up, so I had to group some keywords together into a section. Government was independent, agriculture included “irrigation”, while architecture included “temple” and “Philae”. I wanted to visualize the instances of these sections in a way that was different than a bar graph.

Bar graphs are efficient in showing one dimension, which is counting instances. However, I added another dimension to the data visualization by showing overlap ofinstances between two or three keywords as well. A lot of important keywords were related, which stems back to my research topic. In order to visualize my data, I used Adobe Photoshop to create a landscape of rectangles that would clearly show a size difference that was based on the number of instances. I could not use squares or the exact same ratio rectangles because the number of keywords I queried for would not fit visually. In order to adapt to this issue, I made the equation:

Instances × 1000 = size of rectangle (in pixels)

Even though the instances was multiplied by 1000, this did not mean this was the height and width of the rectangle. 1000 was my starting point, however it was often reduced to cater to size restrictions. If I cut 1000 in half, I would multiply the number of instances by 2 so that the shape has the exact same number of pixels still. This effectively showed the number of instances of each term despite occasionally having different ratios. Most shapes had a width of 1000, 500, 333.3, 250, or 200, all of which were easily divisible of 1000. This meant the heights of shapes were commonly instances multiplied by 1, 2, 3, 4, or 5.

Results and Data Visualization:

The complete results are following:


Each shape represents the number of instances. A scale is in the bottom left corner for reference. Some of the major words feature their counts for viewers to understand. Any square where a “+” is present means both or all of these terms are present in a section. Because “Assouan” is the main keyword, despite it being the third largest, I gave it a textured design to help it stick out more. Shapes with a blue or similar background are related to government, yellow, gold, and orange is agriculture, greens are related to architecture, while pinks and purples are unrelated to any of those three major topics. Inside the “Assouan” shape is “Assouan Dam”, where inside that is what other keywords are combined.

The main data that was collected is nearly impossible to read without zooming in. The following zoomed images show “Assouan Dam” in relation to “Assouan”, as well as the “Assouan Dam” topics alone.



  • Section Agr. refers to (Assouan Dam + Agriculture + Irrigation)
  • Section Arch. refers to (Assouan Dam + Architecture + Philae + Temple)
  • Section Other refers to sections that do not include any of the following terms

“Assouan Dam” was referenced 117 times in the Gazette in the available issues. 114 of these were related to the three main topics, with only 3 being unrelated to any of those. There were no instances of “Assouan Dam”, “Architecture”, and “Agriculture” all in the same section. However, there were 3 instances of “Assouan Dam”, “Government”, and “Agriculture” in the same section, hence the shape between the two. One of the sections that was in the “Other” category was an obituary to a deceased Assouan Dam construction worker. The other two both were broadly about water levels. I did not make this its own section because it was not including statistics and it just was not relevant in the general scope of the dam throughout the Gazette. There were also only two instances, so it just did not seem appropriate to give its own section,and it fits nowhere better than “Other”.

What I found most interesting was how close the three major topics were in the numberof instances. This really shows that there was not one topic that had more significance than another in the Egyptian Gazette. What it also shows is how significant the dam is to the Gazette. Over the span of about 38 months, it appeared in 117 sections. If it was only mentioned in one section per day without multiples, this would mean it was brought up in 117 days during roughly a 1,140-day period. I do not know exactly how much attention a topic needs to be significant, but I would imagine seeing the dam in about 10% of issues over 4 years is pretty significant. Although I did not collect the dates of all these instances, I noticed that a good number of sections that fell under the “government” section came in late 1907 and in 1908 after the flooding occurred. This makes me believe that in the issues that come after what I have access to are likely going to include “government” sections at an increased rate. Most of the sections talked about how the government responded to the flooding, or people writing editorials complaining about the lack of help the government offered before and after the flooding.

A potential future project could be looking at how these instances progressed by year in order to see if there were trends. Perhaps using this chart and making the solid color a gradient based on how many instances each topic saw in a year or month even. Another question I have is wondering if the most significant topic connected to Assouan was not the dam. The most common results are through advertisements and ferries, but I would like to compare the dam topic in its entirety to another topic if there is a major one present.