Kneading
Working a dough mechanically to develop its gluten network into something elastic, extensible and able to hold gas.
Kneading applies mechanical work to a hydrated dough, aligning and cross-linking the gluten network until it is strong enough to trap fermentation gas and extensible enough to expand as that gas grows.
It is the most over-specified technique in baking. Instructions give times — eight minutes, ten minutes — that describe one flour at one hydration worked by one pair of hands. The dough itself gives a far better signal, and it is the same signal regardless of the route taken to it.
Development also happens without kneading. Time alone does a substantial amount of it, which is the basis of no-knead methods and of the autolyse. Kneading is a way of arriving faster, not the only road.
What it is for
To develop the gluten network to the point where it can retain fermentation gas without tearing, and stretch as that gas expands.
RecordedStanley P. Cauvain, Springer
How to tell it has worked
- The dough surface changes from rough and shaggy to smooth and slightly glossy.
- The dough becomes noticeably springier and resists stretching more than it did.
- A small piece can be stretched thin enough to pass light through before it tears — the windowpane test.
- The dough pulls away from the work surface and holds a shape rather than spreading.
What it changes
| Property | Effect | Why |
|---|---|---|
| Structure | Increases | A more developed gluten network holds gas without tearing. |
| Volume | Increases | Gas retained rather than lost translates directly into rise. |
| Crumb texture | Changes | A stronger network produces finer, more regular cells; underdevelopment gives a coarse, irregular crumb. |
By hand and by machine, and why the timings do not transfer
A dough hook and a pair of hands deliver work at completely different rates, so a time that is right for one is meaningless for the other. Machine kneading is also much faster than most people expect and much easier to overdo.
The difference that matters is not speed but FEEDBACK. Hand kneading reports continuously — the dough changes under the hands, and the change is obvious. A mixer hides it, and the dough is only inspected when it is stopped. This is why over-kneading is difficult by hand and routine by machine.
Machine kneading also warms the dough. Friction from the hook adds heat, and a dough that finishes warm ferments faster than the recipe assumed, which shows up later as a proofing problem nobody connects to the mixer.
The reliable approach with any machine is to stop early and check, then run in short increments. The dough is the instrument; the timer is a rough guide to when to look at it.
Under-kneaded and over-kneaded look different, and both look wrong
UNDER-KNEADED dough is shaggy and slack. It tears rather than stretches, will not hold a shape on the bench, and bakes into a dense loaf with a close, uneven crumb. It is the commoner of the two by a wide margin, and it is entirely recoverable — keep going.
OVER-KNEADED dough is the opposite and is much rarer. The network has been worked past the point where it holds together: the dough goes slack, sticky and shiny, loses the springiness it had, and spreads rather than standing. It is essentially not recoverable, which is why the advice to stop early and check matters.
The two are distinguishable by springiness. Under-developed dough has not yet become elastic; over-developed dough has been elastic and stopped being. A dough that resists and springs back is somewhere in the correct region.
The windowpane test, and when it misleads
Stretching a piece of dough until it is translucent without tearing is the most reliable single test of development, and it is not universal.
A wholemeal dough may never pass it, and not for the reason usually given. The gluten there is under-hydrated because the bran has taken the water, so the network is genuinely weaker however long the dough was worked — the popular explanation that bran physically cuts the strands was tested with inert particles of the same size and found to be a minor effect. A very high-hydration dough is difficult to test by hand at all. In both cases the test reports a failure that is not there.
Use it on white doughs of moderate hydration, where it is excellent, and fall back on how the dough handles elsewhere.
Related
Used in
A moderate-hydration white dough is the case where conventional kneading and the windowpane test both work well.
An enriched dough needs development despite the fat.
Developed further than most home doughs are taken.
A strong well-worked dough is what can be stretched thin.
The science behind it
Kneading is one route to development among several, which is only obvious once the underlying process is understood as separate from the technique.
Equipment needed
Optional rather than required, and it introduces the possibility of over-development that hand kneading rarely reaches.
Glossary terms defined here
Windowpane test
Defined against kneading because that is where it is used and where its limits matter.
Products that depend on this
The dough has to be developed enough to survive repeated rolling and folding without tearing.
A lean dough has nothing but its gluten network holding it, and that network has to be developed — by working, by time, or by both.
Extensibility enough to stretch thin without tearing is developed, not bought.
A very stiff dough with little water in it will not develop by time alone. This is the product where kneading is least optional.
The dough must be developed BEFORE the butter goes in, because fat coats gluten proteins and prevents them joining. This ordering is the whole method, and no pantry supplies it.