My article the Hundred Syllable Vajrasattva Mantra has been published in the Western Buddhist Review, Vol.5. This article draws together some of the ideas which can be found on the Visible Mantra Vajrasattva page, but goes further to consider why the received version in the Tibetan tradition is different from the Sanskrit version in the Triratna Community Pūja Book, and to consider issues of authority in Buddhism.
My Teacher Sangharakshita received the mantra from Dudjom Rinpoche and part of the reason for writing the article was a brief dialogue we had about the version of the mantra to chant. Sangharakshita was very keen that we pronounce the mantra as he received it.
The Sanskrit version was an intriguing reconstruction created by Dr Andrew Skilton in 1990 (who was at the time a member of the Order). From a mantra which seemed like the usual collect of mantra words, arranged with no regard for grammar he managed to find a series of well-formed Sanskrit sentences. His main procedure was to place word breaks differently.
Far too late to affect the article I discovered the mantra in the Sarvatathāgata-Tattvasaṃgraha and with my friend Maitiu O'Ceileachair established that it was the earliest version of the mantra in the Chinese Canon, which makes it likely to be the earliest version full stop. We compared the Chinese versions, with two Tibetan, and two Sanskrit manuscripts, and realised that Skilton's reconstructed version was largely accurate. So I have identified a problem of authority. Clearly for many Buddhists it is more important to recite a mantra as it was given by a teacher; while for others it is important to recite it as it originally appears in the texts. A third group are concerned with having accurate Sanskrit above other conditions, and are happy to correct both teachers and texts where obvious errors exist. The Hundred Syllable Mantra offers scope for all three groups.
In addition I make some observations about the types of changes that can be observed in the Tibetan. I was somewhat surprised to realise that the most plausible explanation was that at some point a Tibetan misread a Sanskrit manuscript, as opposed to mishearing, or simply mispronouncing the mantra. This point is related to how Sanskrit is written and in particular how sandhi affects word breaks. I recognised the mistakes as one's that I often make as a neophyte Sanskritist.
I avoid coming to any firm conclusion regards authority - rather than trying to prescribe, I am trying to describe the present situation and set out the arguments for both sides. I do have an opinion, but I wanted mainly to make clear what the issues are as a way to stimulate discussion within the Triratna Order.