Paul,
In the strictest and most basic sense, the classic autolyse as Prof. Calvel described it entails combining only the flour and water. In his book,
The Taste of Bread, at page 31, this is what is said about autolysis:
Autolysis is the slow-speed premixing of the flour and water in a recipe (excluding all the other ingredients), followed by a rest period. The other ingredients are added when mixing is recommenced [...]. During experiments in 1974, Professor Calvel discovered that the rest period improves the links between starch, gluten, and water, and notably improves the extensibility of the dough. As a result, when mixing is restarted, the dough forms a mass and reaches a smooth state more quickly. Autolysis reduces the total mixing time (and therefore the dough's oxidation) by approximately 15%, facilitates the molding of unbaked loaves, and produces bread with more volume, better cell structure, and a more supple crumb. Although the use of autolysis is advantageous in the production of most types of bread, including regular French bread, white pan sandwich bread and sweet bread doughs, it is especially valuable in the production of natural levain leavened breads.
In practice, the amount of flour used for the autolyse rest period can be all or a good part of the formula flour. But the salt and yeast are added later. Craig has already touched upon some of the reasons for their omission but I can provide you with further discussion on this should you wish. But, with few exceptions, which I will touch on below, the classic Calvel autolysis does not include salt or yeast in the dough.
In terms of the duration of the autolyse rest period, I have read from about ten minutes (usually in a home setting) to several hours, even including overnight. This seems to be one of those cases where some people conclude, erroneously, that if a little is good, more, even much more, must be better. As you can see from Reply 15 at
http://www.pizzamaking.com/forum/index.php/topic,3220.msg74624.html#msg74624, the autolyse rest periods used by Prof. Calvel himself were quite short, and that was for dough batches in excess of 75 pounds. Also, as noted in Reply 489 at
http://www.pizzamaking.com/forum/index.php/topic,576.msg30141.html#msg30141, Evelyne Slomon suggested a five-minute autolyse rest period in a home setting. And her post was in the epic Lehmann NY style thread.
The duration of the autolyse rest period is important because that is where the abovementioned exceptions come into play. Although Prof. Calvel said not to add the salt and yeast to the dough before autolysis, from what I read he came to accept the addition of yeast into the dough during autolysis if it would not start to acidfiy the dough during the rest period used. That applied both to commercial yeast and sourdough cultures. So, in your case, if you intend to use a brief autolyse rest period, you should be able to add the yeast before autolysing the dough. However, I don't recall ever reading that salt could be included in the dough during the autolyse rest period. If I had to guess, it was because salt inhibits certain enzymes that soften the dough during autolysis and can result in an undesirable strengthening of the gluten matrix. However, that said, it is also important to note that essentially any dough that is allowed to rest can have an improved texture. Norma had been demonstrating this recently with her high-hydration Detroit style doughs, where after mixing everything together, including the flour, salt and yeast, a rest period improves the texture and handling qualities of the dough (see, for example, Reply 1349 at
http://www.pizzamaking.com/forum/index.php/topic,21559.msg236901.html#msg236901).
It is up to you as to whether to include the yeast and salt in the dough during the rest period. Whether you are a purist about autolyse or not, you are likely to get some benefits from the rest period, just as Norma demonstrated. However, should you decide to add the yeast and salt later, then I would tend to agree with you that holding back part of the formula water to prehydrate the yeast is perhaps a good way of distributing the yeast more uniformly into the dough. Otherwise, you may find that the fermentation process is slowed by adding the yeast in dry form, as I have discovered in tests that I have conducted. As for the salt, it is common in certain dough applications (some of which really irritated Prof. Calvel) to add it late in the process and, as such, it appears to be readily distributed into the dough without first dissolving it in water.
All of the above is directed to autolysis. As Craig noted, complying with the above suggestions might not solve all of your problems. But they shouldn't harm your dough in any way. It may become more a matter of whether you like the results of using autolysis. Not everyone uses autolysis and not everyone likes it.
Peter
