A bit out of my depth with this – a little technical, but here goes – I apologise in advance if I don't explain it properly.
While I don't subscribe to the 'shortened attention span' theory, I do believe that one of the major problems of listening to ballads is connected with the length of some of them.
Following the making of the Radio Ballad, 'Song of a Road' MacColl and Charles Parker, with the help of a group of schoolteachers they were working with, began to take an interest in comparing concentration lengths in connection with vernacular and educated speech.
They came up with the idea that, after a certain period of time listening to speech (in the 'Road' Ballad for instance) delivered in a certain manner, the brain/ear connection appeared to become less efficient, causing the listeners to lose concentration. The length of time appeared to vary with different types of speech.
With the help of the schoolteachers they experimented by playing various recordings to young teenagers of, say, some of the technicians working on the M1 motorway, planners, draughtsmen, etc., and found that the attention span of the listener was much longer when listening to, say a bulldozer driver from Connemara or a concrete layer from East Anglia, than it was to a surveyor or inspector or manager.
Where they put some of it down to the content of the speech, they found that much of it was due to the manner of delivery, dynamic, change of volume, tone and effort, sometimes very slight, as used in vernacular speech, compared with educated/trained speech which tends to be all on one level; (several examples of this in Song of a Road).   
MacColl became interested enough to apply some of the findings to his own singing, particularly of the longer songs – the ballads
He reasoned that if you have a ballad say of twenty-plus verses with a four-line tune – A-B-B-A structure (one of the most common) for instance, you had to do something with that tune to keep the attention of the listener; he called it "making the ear work".
His solution was to divide the ballad up into sections and select places where he would make changes, subtle ones, but enough to retain the listeners' attentions. These would take the form of slight structural variations to the tune, small bits of ornamentation, little alterations in dynamic or tone… a whole bunch of devices that served to prevent the listeners' attention drifting.
He was aware that this could become an exercise in technique rather than interpretation, so he argued that it should always be related to the contents of the narrative, using it as emphasis of a part of the plot, such as time or circumstance or location changes.
If overdone, it could become theatrical; for me (especially in his earlier recordings) it occasionally did, but most of the time it worked – for my taste.
It was argued by some that traditional singers never resorted to such techniques, though we only caught the dying embers of our tradition, so we really don't know what the singers did at its height.
One of the things that we have certainly lost is the familiarity with listening to the unaccompanied voice.
The lack of variation, both within a song and throughout a whole set (or even a whole night) of singing is, I believe a major problem in the revival.
Jim Carroll