Can I find the direction to land when I am in the middle of the ocean and can not see any land? This is a theoretical question. Any method that doesn't use over-technical methods such as smartphones and GPS will do. Example: Ocean current direction, wind direction, fluctuation in temperature, etc.
Your best bet without modern tools would be a sextant and the proper maps for celestial navigation. This is a time-tested and old method of navigation no longer in regular use, but it probably doesn't meet your "no over-technical methods" criterion as it requires a tool, maps and knowledge of its use. If you only had knowledge of the constellations and a few stars (e.g. Polaris, the Sun) you could probably figure out what region of the globe you were on and make a best-guess as to the direction of land.
You might be able to do a similar guess with surface winds, ocean surface flow and temperature to get a rough location but if any synoptic conditions are masked by storms or local anomalies then this will likely not work. There are some semi-presistent weather features (e.g. the Bermuda high) that you could use along with knowledge of the general circulation that could potentially help you out. You could make similar estimates with ocean current and temperature (if you have equipment to measure this), but I think this would only provide regional benefits and may not work everywhere you could find yourself. These would also be marred by tropical convection (e.g. in the wake of a hurricane sea surface temperatures will be cooler than normal) and other storms.
An experienced Polynesian navigator would be the best person to answer this question but as a qualified ocean yacht master and having taught Coastguard courses including navigation, I believe there is some evidence to show that within the trade wind latitudes, where the winds are often steady and constant for extended periods of time; the refraction patterns in the waves which result from waves combining after passing islands, may be detectable beyond the horizon. How far that is depends on the height of eye of the navigator and the height of the island. (the horizon at sea level in a typical Polynesian canoe would be around 4 Nm.) The width of the island as well as the wavelength of the refracting waves would also affect how far the refraction pattern travels or extends beyond the island. However, I believe Polynesian navigators may well have observed the wave motion and steered their vessels in the direction of the "calm" or flatter wave areas, thereby leading towards a not too distant island. But in my experience, observation of wave motion or direction is very difficult to view from the deck of a small vessel, so I do not believe it is a method that would be useful or possible beyond even as far as 50Nm from even a large islands such as New Zealand. Besides which, the Polynesian catamarans, are not able to sail close to the wind so following a windward course would require a lot of tacking, to and fro in order to stay in sight of the calmer sea pattern.
The polynesians used primitive star charts, which is OK for latitude, but tells one nothing about longitude. The latter is the age-old problem that faced sailors for centuries, if not millennia, and which cannot be resolved without sophisticated instrumentation. The Polynesians probably supplemented their navigation with other clues, such as bird migratory tracks, current temperatures and strengths, cloud frequency and, as Tony says, wave refraction features.
If you know your current position, and the position of the land youʻre headed toward, then you only need to be able to determine direction and keep track of your progress (perform dead reckoning). You can determine direction from some key stars (the sun rises and sets near east and west, and the exact direction depends on the time of year). The North star is always north, and Mintaka in Orionʻs belt rises due east and sets due west from every latitude on Earth. Other starsʻ rising and setting points depend on your latitude to varying degrees. You can roughly check your latitude by the height of Polaris, the North Star, above the northern horizon, or by the height of the upright Southern Cross above the southern horizon. There are also certain pairs of stars that point very nearly due north or due south when they are stacked vertically. For example, Dubhe and Merak when aligned vertically are nearly directly above due north. They also act as a guide to locating the north star.
If you donʻt know your position, or the position to land, then you need some kind of indicator from the land itself. In the trade wind belts, there is often a stationary pile of clouds over high volcanic islands, whereas most cumulus clouds float freely downwind. It is alleged that clouds directly over coral atolls with central lagoons may sometimes reflect a green color from the lagoon. Seabirds such as the brown noddy and fairy tern that are known to sleep on land at night can be used to locate land, since they head out to sea in the morning to fish and fly back at night to sleep. That gets more complex when they make multiple fishing trips to feed their young. Lines of rubbish and organic matter (leaves, sticks) may indicate land is nearby. If you know the prevailing direction of the swells or can calibrate them with the rising or setting sun or recognizable star, swell directions typically donʻt change quickly in under 24 hours. If there are multiple swell patterns with the exact same period (time between swells), that may indicate that one of the swells is a reflection off of an island of the other swell, or a refraction of the original swell bent by the shallow water around an island. This was a mainly Marshallese island technique. They were noted for their wave pattern stick charts, which were apparently more used for studying and teaching than actual use during a voyage.
We, The Navigators by David Lewis chronicles much of this.