Why is it that Peter Reinhart most often uses a small amount of sourdough starter in a recipe and Nancy Silverton a large amount?
Does Peter use a small amount of starter to make a medium-sized amount of an intermediate thing ('poolish', 'biga', 'sponge', there are many names for these things) which he then adds to the final dough?
You never want to dilute your starter TOO much, since you'll wind up creating a hostile environment for the yeasts and bacteria, so in the final dough you usually want 20 to 50 percent "starter" of SOME sort. If you take an active "ripe" starter (that is: any mixture of flour and water with an active and mature population of yeast and bacteria that has almost eaten up all the sugar in the mixture) and dilute it, say, 20 to 1, you're going to get in trouble (I think). 5 to 1, 3 to 1, no problem. 10 to 1, I don't know -- you're getting in to a grey area, I think.
You can build that final amount of "starter" or whatever you call it, for your final dough, in stages. This has the effect of digesting various amounts of the flour by various amounts (you have a mixture, say, of 24 hour old flour with 12 hour with 4 hour or whatever), or you can simply start with a large amount of whatever your starter is that's in a jar etc. If you do the latter, then you only have flour of two ages in your dough, which will alter the way the flavors play out.
Either way will work, though!
Now that I think it over, I'm not so sure 20 to 1 is that iffy. I often use something like 2T ripe starter to 1 cup flour and 1 cup water, which has got to be something like 15 to 1 by weight.
Anyways, the point is that there's SOME ratio at which your culture will struggle in an inhospitable environment, and the lower the dilution, the more quickly the culture can take hold and grow.
At high dilutions you're getting a long soak time (the flour is simply "wet" for a long time) before your poolish/sponge/biga/levain gets ripe, which alters the chemistry and flavors to a degree. You're probably also encouraging different subpopulations of your culture more or less (maybe the lactobacilli are happy at high dilutions and the yeasts are sad and slow, or vice versa) so you're getting different kinds of fermentation and fermentation products, depending on how much you dilute your culture.
My two favorite methods of leavening with natural yeast both build an active culture thing that's 20 to 40 percent of my final dough:
2T liquid starter + 1 cup water + 1 cup flour: let rise overnight, add 1 to 1.5 cups more liquid and flour. This is a high-dilution sponge, going straight into dough.
1T liquid starter with a little water (1/4 cup plus or minus) and enough flour to make a stiff dough. Ripen overnight. Break up into a little more water (1/3 cup more or less) with enough more flour to make a stiff dough. Ripen. Make bread. This is a low-dilution "biga" style thing, built in 2 (or more) low-dulition stages, which then goes into the dough.
The final thing that goes in the dough is more or less the same, but in the first case it's a cup of flour that's been soaking all night while a small population of critters clawed its way up to a big population. In the second case, some of the dough spent the night wet, and some has only spent a few hours, and the critters never had to fight that hard since their environment expanded more slowly.
The results are definitely different, and I'd be hard pressed to describe in detail what the differences are. But, they're there!