dom4j: flexible XML framework for Java

Here is an XML document

<?xml version="1.0" encoding="latin1"?>
<taglib xmlns="http://java.sun.com/xml/ns/j2ee">
<tlib-version>0.1.1</tlib-version>
<jsp-version>1.2</jsp-version>
</taglib>

I try selecting the taglib node:

xmlDoc.getRootElement().selectSingleNode("/taglib");

But all I get is null.

If I use prefixes I can get the taglib

SOURCE XML
<?xml version="1.0" encoding="latin1"?>
<tld:taglib xmlns:tld="http://java.sun.com/xml/ns/j2ee">
<tld:tlib-version>0.1.1</tld:tlib-version>
<tld:jsp-version>1.2</tld:jsp-version>
</tld:taglib>

CODE
xmlDoc.getRootElement().selectSingleNode("/tld:taglib");

With Xalan I can use /:taglib to get the taglib in the first
scenario (We already talked about it in previous posting)

It was not standard compliant syntax but my question is
how can I select the taglib in according to the standard.

I tried /taglib[not(contains(name(), ':'))] but that also
returns null.