Autolyse link repair
I discovered today that there is a broken link that probably exists in a number of places on TFL. The link points to an extract from Advances in Baking Technology, B. S. KAMEL AND C. E. STAUFFER (1993 seems to be the most recent edition) Page 62:
A Google Books link here from what may be an update to the original. Scroll down in the Table of Contents and click on the Page 62 link.
This link is now correct. The target article is a short read (one page) that makes the point that the purpose and the mechanism of autolyse is to allow the water to be fully absorbed by the flour. The method involves an initial mixing operation that develops enough gluten to produce a "clear dough" which from other research means a dough that pulls off the side of the mixing bowl (in the example case when mixed using a spiral mixer).
You can easily show that autolyse is a physical process rather than a biological process by observing the relatively small difference between doing it cold and doing it warm.
Doc
Actually, that is not a usable link for many folks because you need an account on Google Drive. I don't have one, I'm not going to set one up, and I'm sure I'm not the only one.
I hope you have some other link for this work.
I have an account for ChatGPT and I asked it for a summary of the article. Hopefully this will not be a copyright violation. Here's the transcript:
Prompt
=======
please summarize a short article about the autolyse process in bread baking that appears in extract from Advances in Baking Technology, B. S. KAMEL AND C. E. STAUFFER
ChatGPT
========
Certainly, the autolyse process in bread baking, as discussed in the extract from "Advances in Baking Technology" by B. S. Kamel and C. E. Stauffer, is a crucial step in breadmaking. Autolyse is a technique that involves mixing flour and water and allowing them to rest for a period before adding yeast and other ingredients. During this rest period, gluten development begins without the addition of yeast or other dough enhancers.
This process typically lasts for 20-30 minutes and has several key benefits:
Gluten Development: Autolyse encourages the formation of gluten, the protein responsible for dough structure. This leads to improved dough strength and elasticity, making it easier to work with and resulting in a better crumb structure in the final bread.
Enzyme Activity: Autolyse also allows enzymes present in the flour to start breaking down starches into simpler sugars and proteins into amino acids. This enzymatic activity contributes to enhanced flavor and improved fermentation.
Reduced Kneading: Because autolyse aids in gluten development, it reduces the need for extensive kneading, saving time and effort in the breadmaking process.
In summary, the autolyse process is a vital step in bread baking that involves mixing flour and water and allowing the dough to rest. This technique promotes gluten development, enzymatic activity, and flavor enhancement, ultimately leading to improved bread texture and taste while reducing the need for vigorous kneading.
For what it's worth, I went on to ask ChatGPT about the relative merits of including all the ingredients before the rest period, vs only the flour and liquid, especially for a long bulk fermentation. The responses were OK, but not really that helpful. I think that's because it's one thing to summarize an article, and another to actually know something in detail about a subject.
I could post the transcript if someone is interested.
Thanks Gavin. The process of fixing the link broke the old sharing setting which I have now fixed. You don't need a Google account to view a shared document.
Some caution is appropriate when you read a ChatGPT summary as it knows knothing, can't think, doesn't analyze, and parrots pervasive but erroneous information as it learns from bad information as much as it does from good information. The fact is that many baking schools still teach that autolyse depends on proteolysis to degrade gluten without being able to explain how the biochemistry works or point to any technical literature that substantiates the claim. It is just a wives tale told over and over without anything to support it.
I think Professor Calvel got it right, he just didn't say what others have claimed he said. It may be that European flour in the post WWII period was not easy to make into crusty bread (think baguettes) and autolyse helped strengthen the gluten without subjecting the dough to over-mixing. This is just speculation on my part, but it is certainly true that the varieties of wheat available today make better bread than the wheat that was being grown in Europe in 1956.
Some caution is appropriate when you read a ChatGPT summary as it knows knothing, can't think, doesn't analyze, and parrots pervasive but erroneous information as it learns from bad information as much as it does from good information.
I'm certainly aware of this, and I posted it mainly for interest, not claiming it was accurate. But the summary part seems reasonable, whereas when I prompted it to go further, the results became questionable,
If you knew nothing about autolyse, the ChatGPT output would point you in the right direction for further digging. I thought it did a credible job without making any major errors. The summary of autolyse is "a rest period that allows wheat dough to fully absorb water" and nothing else, yet I have not found any reference that states it that simply.
Yes, and that's a good way to look at the use of tools like ChatGPT. It turns out that crafting the prompt carefully can make a big difference in the results. Instructing it to act as an expert in the relevant field is usually very helpful, and asking for conciseness or brevity cuts through a lot of the verbiage you might otherwise get.
I always thought autolysis was implemented on slow speed with minimal gluten development?
Lance
My take on it has been based on the observation that resting the dough after mixing lets it hydrate and combine, and after a rest the dough is much smoother and obviously has some gluten development. At this point, it's much easier and faster to finish smoothing the dough and further developing the gluten. Why bother to mix and knead for 10, 15, or even 20 minutes, when you can do a rough hand mix in minutes, wait half an hour, and then knead/stretch for a few more minutes and get to the same point?
I've learned that omitting the salt until after the rest tends to make for more extensible dough. So when I expect the dough will be more elastic and less extensible than I want, I wait on the salt, otherwise I add it along with everything else.
Since I usually want a long bulk ferment, I've never thought it made much difference whether I add the starter before or after the rest. If I wanted a very short cycle from mix to bake, I'd balance these factors differently.
Now, what really counts as an "autolyse"? Is it an autolyse if I add the salt and starter/yeast during the initial mix? I don't know. I'm happy to call the quiescent interval a "rest.
Now this works well for mainly wheat breads. High rye content breads are a different world.
If you don't mix enough to get all of the flour wet, it will take longer for autolyse to work since it depends on diffusion for water transport. If you do it cold it will take longer (though not a lot). No knead doughs provide the evidence that gluten will form without heavy mixing, it just takes time.
I am sure that Dr Calvel did a lot of experimentation to optimize the process, but if you can't tell the difference there is none so why not just do it the easy way. I mix in the levain before autolyse because it is easier: levain, water, flour, mix, autolyse, add salt, develop gluten, bassinage to target hydration, bulk ferment until dough is saturated with CO2, divide, pre-shape, rest to relax dough, final shape, proof and bake. That is my short form instructions for making bread. Salt could go in early, bulk fermentation can last longer; do what it takes to make the schedule work for you. Nobody else matters. Use temperature to control timing when appropriate.
You and I are right on the same page.
So long as the flour has fully absorbed the water by the time you say autolyse is done, that must be OK since that is the objective. Calvel's endpoint was apparently enough gluten development to make a "clear dough" which is more than "no gluten development" but might be the same as "minimal gluten development".
Should the term "autolyse" no longer be used to describe the technique? Its connotations to a biological process confuse the fact that it is a physical process of hydration of gluten and starch.
Or, are there enzymatic processes occurring during the rest that would be biologic in nature?
Even a comment on Wikipedia asks whether it should be included in the Autolysis (biology) entry.
Yes, the term autolyse is misleading, but it refers to Dr Calvel's process which seems to work just fine at temperatures too low to energize biological processes. I have pre-chilled flour to 0°F then added 40°F water and mixed until incorporated. The dough was then held at 34°F overnight. That batch made good bread. That is not hard evidence of an absence of biological processes being involved but if there was much biology involved it didn't run very fast and the rate limiting factor was most likely temperature. If somebody has a better word to describe the process I would entertain using it. I think the point is that the effects observed result from near complete absorbtion of the water by the flour and do not depend on any biological activity. I have no desire to change what people do, and I will continue to use it and call it autolyse (until somebody comes up with a clever alternative). It took over ten years for me to become convinced that it was not as claimed and the reference I pointed to was truely enlightening - an industrial process called autolyse that is widely used without any consideration for involving a biological process.
The following are my thoughts:
Certainly, what autolyse means in practice varies and therefore what happens on a chemical level varies also.
In a short time autolyse (< 2 hours) it would be fair to say that not a great deal of transformations occur which are detriment to gluten or inhibit their ability to form a strong (elastic) gluten network (macropolymer).
However, there are considerable many transformations that occur in quick succession as a result of native enzymes, typically of starch and non-starch polysaccharides, lipids and other constituents which have either a direct or indirect effect on interactions with gluten proteins.
The framing from the OP does open the door for revisiting our existing understanding of gluten development.
In respect of that, passive development i.e., not employing active work (energy) to develop the dough is commonly utilised and accepted as an ideological approach in artisan baking.
However, the thinking that a passive approach alone can develop (transform) the gluten network in an equivalent manner to that of mechanical mixing is very much misunderstood.
In practice the two approaches give different results.
Furthermore, the thinking that gluten development is merely a factor of hydration is equally a false summation. The input of energy endorsed through mechanical mixing does in effect invoke a physiochemical transformation which passive approaches can’t replicate.
In part, mixing is responsible for advancing hydration, but it also exerts a mechanical force which is reciprocated by the forming gluten and its ability to physically resist. As this action continues it distributes the gluten proteins into a more ordered distribution (reverse entropy).
Once fully distributed the mechanical input equalises with the resistance accrued by forming protein bonds. At this point gluten can be considered fully developed. Further mechanical work will begin to depolymerise the protein bonds to such an extent that they become unrecoverable (sticky mess).
Utilising passive development allows for the slow natural absorption of water (hydration), however it does not effectively distribute the gluten proteins to form an extensively maximised macropolymer. The missing factor is energy!
Passive approaches alter our (the baker’s) perception of development and mask the otherwise measurable determinate of full development.
Long autolyse processes (>2 hours / <24 hours) do enhance the potential for the effect of proteolytic enzymes that occur in non-acidic environments.
Without invoking mechanical work upfront, gluten proteins occur in smaller aggregates which with their smaller (lower molecular weight) are more soluble and being more soluble they are more susceptible to enzymatic cleavage. In an autolyse the terminals of gluten proteins are cleaved to give free amino acids. Since free amino acids act as precursors to flavour compounds it has been observed that doughs with higher free amino acids (FAA) results in more flavourful baked bread.
Mechanical work also helps to increase oxygen incorporation, and this promotes the formation of very strong disulphide bonds which form the linkages between glutenin proteins thus contributing to the formation of the gluten macropolymer (network).
With all this is mind I concluded some time ago that an autolyse is essentially a reductive phase. That is at least how I see it!
@Micheal - I absolutely agree with your points. The only thing I am pushing back on is the insistence by some that the process we call autolyse (resting wet flour for a while before continuing with making bread) is dependent of proteolytic enzymes degrading gluten proteins. All of the dependence of gluten development on shear forces to break intra-molecular SH bonds and allowing them to reform as inter-molecular bonds is the real process that yields what we know as good bread is the essence of what we have learned from over 100 years of scientific investigation. I am disappointed that baking schools continue to teach that proteolytic activity is the dominant mechanism of autolyse. Your articulation of the details I find interesting. And thank you for pointing toward other avenues of productive investigation. It seems clear that Professor Calvel did not understand you points well enough to explain them in any of his writings.
As with nearly all things - if you want to understand what is happening, it is all revealed at the micro scale.
I remember a few years ago, there was an "artisan" fashion for very long autolyses. I was a bit greener in those days, so I slavishly followed such a recipe; it called for an overnight autolyse at ambient temperature without any salt.
Sure enough, in the morning I went into the kitchen to be presented with a bowl of dough soup - completely unusable. At that moment I made a vow never to do excessively long autolyses ever again.
But presumably that wasn't purely a physical hydration effect of autolyse? I guess there must come a point in time when other enzyme based effects start to take effect.
This will vary depending on many factors, eg temperature, hydration, flour type.
Lance