Nostradamus: what the Centuries actually say
Michel de Nostredame (1503–1566) was a physician, an apothecary and a writer of almanacs. His Prophéties appeared from 1555 onwards and consist of about a thousand four-line stanzas, arranged in groups of a hundred — the Centuries.
How the texts are put together
Every stanza has four lines. The language is a Middle French studded with Latin, Provençal, Greek and words that appear to be invented; anagrams and distorted words are frequent. There is no chronological order — the stanzas are not in the order in which they are supposed to come true, and they almost never name a date.
These three properties — obscure language, no order, no dates — govern everything that follows.
Why every interpretation works
With a thousand stanzas in no order and without dates, a fitting one can be found for any event in history. That is not a reproach to Nostradamus, it is a property of the quantity of text: given enough statements that carry several meanings, a hit is certain.
Then there is the translation. A Middle French word with several meanings is pinned to one of them when it is carried over — and the pinning is done by someone who already knows what the verse is meant to fit. The best-known case is the place name “Hister”, which in sources of the period denotes the lower Danube and was reinterpreted in the twentieth century as the name of a person.
The one test
A prediction is worth something when its interpretation is settled before the event.
That is exactly what is missing. Practically every known hit was matched up after the event. In the few instances where someone tied a stanza in advance to a particular future event, the prediction did not come true — the events announced for 1999 are the best-known example, and one of the few in which the text itself names a year.
What Nostradamus actually did
He was a well-educated man of his time and worked with the means then available: almanacs, astrological forecasts for the year, the doctrine of the great conjunctions. In his lifetime his reputation rested on those yearly almanacs, not on the Centuries.
As evidence of sixteenth-century practice the texts are interesting. As predictions they do not meet the one standard that counts for predictions.
See also: Edgar Cayce and twentieth-century prophecy — the same question, better sources.