This week's book giveaway is in the OO, Patterns, UML and Refactoring forum. We're giving away four copies of Refactoring for Software Design Smells: Managing Technical Debt and have Girish Suryanarayana, Ganesh Samarthyam & Tushar Sharma on-line! See this thread for details.
The answer is the same as the answer to all the other "why is it like that" questions I see around here: it's that way because the designers decided to do it that way.
Now perhaps you meant to ask why the designers decided to use XML. Or perhaps you meant to ask why XML might be a better choice than some other format. But those are very different questions and we shouldn't be made to guess what you meant by your question. So... if you have a specific question, please ask it.
Years ago when servlets first appeared I worked with a pioneering commercial version.
It used properties files to define parameters to the container - what a confusing mess.
Because of the limited ability of the property file concept of name/value pairs to communicate structure, we ended up with multiple files having confusing names containing properties with confusing names. Switching to web.xml was a big improvement.
In my very first J2EE project, back in 2001, we used BEA WebLogic version 5.1. You had to configure it in one giant *.properties file. That was a big mess. In WebLogic 6.0, the configuration was done in XML, which was much more clear.
A *.properties file is just a flat list of key-value pairs. XML has hierarchical tree structure, so that you can organise things in a much better way.