Frosh agrees with and amplifies my earlier analysis of the symbolic dimension of the 11 September events and its implications about the profound long-term dangers of the landscape of symbolic inequality, which today’s global media system represents.

He disagrees however with my concern, raised at the end of my piece, at the particular foregrounding of images of destruction over analysis in the days immediately following the events. But there are two problems with the way he develops his criticism of what he calls ‘the weaknesses of [my] exit-strategies’ from the mediated global endgame that we both fear.

First, he doesn’t mention that we would seem to agree on the first two of those exit-strategies: the necessity for a shift in the agendas of representation of world media organisations, and the implications of that shift for long-term investment and day-to-day news priorities. We should not lose sight of that agreement.

Second, he mistakes my third exit-route – my call for greater caution about an excessive use of and reliance on images at times of great emotional turbulence – for a general attack on media images as such. But my comment - and admittedly I could have made this clearer – was meant not as a general ‘iconoclasm’ (as he puts it), but to be taken within the context of my analysis of the consequences of the unequal symbolic landscape on which we both agree. It is because the power to produce images on the scale, intensity, and with the pervasive reach of the images of the 11 September destruction is so unequally distributed – and it likely to remain unequal for a long time to come – that I argued for more caution about a use of images that simply works with, rather than questions, the contours of that landscape. Which was why I suggested that words that distanced us from the power of those images from the centres of world media production were important.

I wasn’t therefore denying that at the same time, in so far as we have images, they need to come from a greater variety of sources, as Paul Frosh says. In fact, that is exactly what I argued earlier in the piece. But images alone cannot be enough. Nor do I see much practical value in the concept of ‘dissemination’ which Frosh draws from John Durham Peters’ excellent writings on the philosophical aspects of the media process; ‘dissemination’ as an abstract concept does not take us much closer to understanding or adjusting for a situation where the channels of dissemination are so unevenly structured in advance.

What we need in fact is more detailed, specific debate about the ethics of both image and word, and their interrelations. This is the difficult and largely uncharted area that I was gesturing towards at the end of my piece. It would be good if openDemocracy were one place where this debate could develop. For this difficult terrain, it will be important to avoid polarising our positions more than we need.