How do you explain the physics of tacking (sailing to windward, i.e. at an angle less than 90 degrees to the wind direction) to someone who has a little college physics but nowhere near mastery?

I'm not sure anybody's nailed the "plain language" or "College physics" version yet, so here's a try. BTW, when I was actually STUDYING College physics (a long time ago), I was convinced that it was easier to prove that a sailboat could NOT make progress into the wind (that it would go backwards if it tried), than it was to explain how it DID it! (I was already sailing a lot, so I damned well knew that it COULD do it!) The main "trick" in the explanation is that there are TWO vector resolutions, not just one. The more obvious one is probably the keel or centerboard, which pretty successfully keeps the boat from sliding sideways. So it translates ("resolves") any force, from any direction, into either FORWARDS or BACKWARDS motion. (If you like, imagine the blades of an iceboat, or the wheels of a land-yacht -- or put your sailboat on a straight railroad track!!) That doesn't solve our problem by itself, because the "raw" force from a headwind, when resolved into its fore or aft component, is pushing aft. That's why it seems as if a sailboat should go backwards when steered upwind. And indeed, without sails, or with square sails, that's exactly what happens. So we need the SECOND vector resolution, which takes the raw wind, coming from ahead of the beam -- say "10 o'clock" -- and translates ("resolves") it into a net force that's behind the beam, say "8 o'clock". Once we have a net wind force from 8 o'clock, the keel/centerboard will ensure that we move almost perfectly straight ahead. (The sideways component of the "8 o'clock" push is translated into sideways heeling, by the keel/centerboard.) Well, on any modern sailboat, the adjustable angles of the fore-and-aft sails (main and maybe jib or Genoa) do exactly what we need! They turn a "10 o'clock" headwind into an "8 o'clock" push FORWARD! For this vector resolution, you have to remember that the force of the wind is the vector sum of two components, one that's PERPENDICULAR to the sail (which can be a flat sheet of plywood), and one that's PARALLEL to the sail. Since the sail (even a plywood sail) is very non-porous but quite slick to the wind, we can concentrate on the perpendicular and ignore the parallel. With the wind coming from "10 o'clock", we arrange our sheet-of-plywood mainsail so it's aiming at "11 o'clock". The perpendicular component of the wind's force on the sail is now coming from "8 o'clock"!! So, to recap and simplify: Without invoking Bernoulli, or curved sails, a sailboat with one sheet of plywood for a mainsail and another for a centerboard WILL sail to windward, as follows: With the centerboard extended, aim the boat upwind so the wind is coming from "10 o'clock". Trim the mainsail so it's pointing at "11 o'clock". The net force on the sail (from the sail-perpendicular portion of the wind) is pushing from "8 o'clock". The centerboard turns the big sideways part of that force (from "9 o'clock") into heeling (tipping), and turns the smaller forwards part of that force (from "6 o'clock") into forwards motion, towards "12 o'clock". At this point, our super-simple sailboat IS sailing UPWIND! BTW, as the boat starts accelerating straight ahead, it will add its own "bicyclist's headwind" to the true wind, shifting the "apparent wind" angle from "10 o'clock" towards "11 o'clock". If it ever got to "11 o'clock", it would be perfectly parallel to our plywood sail, and there would be ZERO perpendicular force to push the boat forward, so we'd slow down. In low-friction boats -- landyachts, iceyachts, some catamarans, etc. -- that is what limits the attainable speed to windward. In more "normal" sailboats, we get to "hull speed" first, and that's our speed limit. This analysis has not invoked Bernoulli or curved sails. I haven't mentioned Newton, either, but his laws of motion underpin this explanation. The boat is pushed ahead by deflecting air towards "behind" or "aft". Equal and opposite reactions. Curved sails (and foil-shaped centerboards) do it better than sheets of plywood, but they are NOT necessary to sail upwind, in theory or in practice! NEITHER is the "slot" between two sails, though it also helps do it better. BTW, the "High-School Bernoulli" explanation of airplane flight is more wrong than right, and airplanes fly fine with flat plywood wings, too -- or fancy symmetrical-foil wings, where the path around the top is just as short as the path around the bottom. Or inverted. Bernoulli got it right, but the high-school teachers messed up his theory when they simplified it. But that's another answer!

A sailboat can go upwind because of a combination of two forces - One pushes the boat at about 10 o'clock from the bow (imagine drawing this) and one pushing  at 2 or 3 o'clock.  These forces are created by the sails and keel / centerboard. Upwind, the mainsail and jib function like airplane wings, creating lift off the back of the sail "upwards" (10:00).  This force, combined with the lift created by the keel / centerboard through the water (3:00) produce forward motion when the boat is going upwind. If you think of both the sail foils and keel foils as airplane wings and diagram where the lift occurs, you'll be able to see it clearly. As Martin said above there are all sorts of things that go into the calculation for optimal upwind speed, but the basic principles of foils and air or water apply.

Downwind, the wind pushes a sailboat alone.   Upwind, the sails and keel generate lift in accordance with the Bernoulli Principle [1]. While tacking, a sailboat transitions from being a lift generating and forward moving vessel to one that has effectively stalled (like an airplane), during which time, the sails and rig create drag and slow the boat -- the sailboat uses it foward momentum to transition from one tack to another, where the sails can fill and reuse generating lift. [1] http://en.wikipedia.org/wiki/Bernoulli's_principle

Only after reading all of the comments here did I realize that there is a slight misuse of terms involved. A sailing vessel is said to be "beating" to windwards when sailing against the wind (be it at an angle to the wind direction). Tacking is what a crew will perform with the vessel when changing from one "tack" to another (ex. tacking from starboard tack to port tack). Beating can also be defined as making positive VMG (Velocity Made Good) against the direction of the wind. So, putting that record straight, and acknowledging what has been said here about the sailing vessel using it's speed and inertia to master the dificult task of tacking, let's move on to the process of "beating" or in general sailing with laminar flow around the sails and the physics involved in actually generating this speed. In my long life as a competitive sailor and an engineering geek I have always sought proper explanations to how a sailing vessel actually works. I was for a long time fed the usual misconception that a (pair of) sails works like the wings of an airplane and that (and here's where the misconception comes in) such a foil generates lift because there is a longer distance "over the top" of the foil than under it (because of the asymetric form of a classic set of airplane wings you see). The question that has always crept around in the back of my mind is. But, but, a sail isn't like that, is it? The answer is, no it isn't. A sail typically don't have a longer way for the fluid to go on one side of the foil (allthough it has long been established that a sail is way more efficient if it has this configuration, thus wing sails, see f.ex. Vestas Sailrocket). It wasn't until I found the book "The Symmetry of Sailing" by Ross Garrett (http://www.amazon.com/Symmetry-Sailing-Physics-Yachtsman/dp/1574090003) that the penny dropped for me. It's a revelation and more yachtsmen should read it.   It's full of rather complex theory but the upshot of it all is that any foil in a moving fliud will generate lift if it has angle of attack (even a piece of perfectly flat plywood will). The angle of attack sets off what is called downwash in aeronautical terms (downwash because it works down towards the ground in the aeronautical case). Following this event a trailing wortex is created and separates from the trailing egde of the foil. Now, positive vorticity cannot be created in fluid without stirring the fluid (which we are not, look at in from a relativistic approach, the foil is standing still, the air is moving around it) so an equal, in strenght, but opposite in direction vorticity must therefore be created. This opposite vorticity, as it turns out in the experiments Garrett refers to and has participated in, is established around the whole foil. What Garrett now says is that a sailplane with an angle of attack to the fluid (air) it is moving in is "transformationally equivalent to a Flettner Rotor" (http://en.wikipedia.org/wiki/Rotor_ship). Read the book to see why.   So, to sum up so far. What we have now is a Flettner Rotor in all aspects, except of course the trivial fact that, relative to the boat, the rotor(sail) is standing still and the fluid system is rotating with the vorticity around the rotor(sail). Makes your brain twist a bit yes? ;) A Flettner Rotor generates lift (perpendicular to the motion of the fluid) because the rotation shifts momentum from on side of the rotor to the other (sides beeing relative to the separation point of the rotor without the rotation of the rotor). This is called the Magnus effect (http://en.wikipedia.org/wiki/Magnus_effect). Shifting momentum from on side to the other means that fluid is now moving faster on one side of the rotor (sail) than on the other and here's where Bernoulli and his principle and/or Newtons law on motion comes in and dictates that there should be a pressure difference between the two sides of the sail (rotor). Now you can insert all that has been said above about the direction of the force created by this pressure difference and the effects it will have on the boat, except of course for the discussion regarding the underwater forces that keeps the whole system in balance (again, read the book). The rest is history and pure fun ;) PS I agree that this would be easier to explain on a whiteboard or a piece of paper, so I think I miserably failed the task originally set up, which would be explain it in a simple way. Sorry about that :-| It's just not that simple. Then again, that's what makes it fascinating.

I think the easiest way to think of it is that the mainsail is just like an airplane wing stood up in the air.  The same way that the airplane wing provides lift, so does the sail.  The keel/centerboard translates that power into forward movement (by keeping the boat from "sliding") sideways.

Let's keep it simple (on purpose), shall we: 1) both keel and sails create "some lift" 2) lift is created when a foil moves relative to a fluid, be that air or water. Not all foils are created equal, some work better, some worse. A traditional sail is not the optimal foil but it works. 3) at an angle <90degree to the wind the sails create lifts (if properly trimmed) with two components, one forward of the boat, one abeam to it and to the lee. 4) by the same physics, the wing keel generates lift moving in the water, one component towards the back of the boat due to friction, one abeam to windward 5) combine (3) and (4) and the resulting forces moves the sailboat towards the wind at <90 degree from it.Not all keel wing are created equals, some will do a better job at compensating the push to lee from the sail, keeping the sailboat from moving sideways. There is a lot more, as many have wrote, but this is the simple version. Note: as the sailboat moves towards the wind, the wind speed across the boat itself increases since you are moving (in part) against the wind. This in turn creates more lift. Resistance from the water and hull speed is what prevents a sailboat from going ever faster against the wind. If you read about it you will discover that in recent years maritime engineering has allowed to build vessel that can actually move towards the wind at a speed faster than the wind itself. Note that such boat are still tacking back and forth, it is just that they are so fast on each tack that the combined velocity towards the wind is faster than the wind itself.

I'm going to give a big '++' to Martin's answer, but also add that - also critical to upwind performance is the skipper growling "Boatspeed ... boatspeed ..." alternating with telling the helmsman he's pinching, and asking the trimmers to fool around with the Cunninghams because skippers never seem to know for sure what the Cunninghams do. (grin)