Before discussing navigation in the Odyssey it is necessary to solve the problem of Homerís system of orientation and explain his terminology in relation to winds.
Much has been written about winds and orientation in the Homeric poems, but the question has been treated unsatisfactorily because it has not been dealt with sufficient breadth of view. One must consider the problem both from a theoretical and from a practical point of view. From a theoretical point of view one must consider Homerís cosmogony, and from a practical point of view one must consider the regime of winds in the Aegean and the Ionian sea.
If one faces the problem as a history of the rose of winds, as it has been done by Rehm, the problem cannot be solved. Rehm and others have assumed that Homer knew a rose of winds with four directions which was later developed into a rose with eight directions in Aristotle, and then became a rose with twelve directions in Timosthenes. The fact is that Homer knows four directions, and these are in a way perpendicular to each other, but they do not intersect at a center of the horizon. Homer does not have the conception of the horizon as a circle that can be moved to various points of a sphere. For him the earth is a flat area included between the rising point of the sun and its setting point. These points are believed to be fixed, so that, for instance, it is conceived possible to approach the very point where the sun sets. The position of East and West is determined by the rising and setting of the sun, but these points change during the year. They are ca. 30 degrees north of true east and west at the summer solstice, and they are ca. 30 degrees south at the winter solstice.1 Of course we take as east and west the rising and setting points of the sun at the Equinoxes. But the idea of taking as standard the equinoctial rising points did not occur to the Greeks up to the time in which they began to consider the equinoctial day as a standard of time. It is only about the time of Aristotle that the Greeks began to use the division of the day into twelve hours equal to the twelve hours of the night. The length of the hour was changed according to seasons of the year; if a standard hour had to be used, it was found convenient to use the length of the hour at the winter solstice, that is, the shortest one that could be fitted into any day of the year.
A similar problem existed for the men of the time of Homer in relation to the determination of east and west. The east could be set either at the rising point of the sun at the winter solstice or at the summer solstice. These are the points that are obvious to any observer, whereas the rising point at the equinoxes is a geometrical point which becomes significant only to the person who considers the problem theoretically. It became relevant only when there was felt the need to determine an hour of equal length for the night and the day: for this reason Aristotle and other scholars call our east and west respectively equinoctial (ishemerinos) rising and setting. For Homer east and west (that is Euros and Zephyros) are the points at which the sun rises at the summer solstice. There are many reasons for this. First of all, the sun rises and sets north of our east and west line during the period important for navigation and the period in which sunrise and sunsent can be more readily observed. In the second place the sun after rising always moves to the south, so that the course of the sun is conceived as an arc (not as a circle as it is for us) with the apex directed towards the south and the two points towards the north. The sun is conceived as rising and setting to the north and moving to the south during the day. for this reason the Aethiopians, who live at the south, are said to be baked by the sunís heat. If one superimposes a Homeric conception of the world on a conception of the horizon according to our standards, one gets a picture of this kind:
The rising and setting points of the sun are fixed, and there are nations that live next to them. Where the sun goes below the level of the earth is the land of the dead, Erebos, in which runs the Oceanus which is conceived as an infernal river. The fact that the sun for some part of the year rises and sets at a point further to the south could be easily explained on the basis of the experience of the sun rising and setting behind a mountain. If the sun rises or sets behind a mountain it does so at a point more to the south than the normal one. Therefore the rising and setting of the sun at and around the winter solstice could be easily explained by assuming that the sun was keeping itself concealed for a part of its course. Therefore the movement of the sun at the winter solstice would be described in this way:
The very fact that the sun is at a lower height at noon would confirm the opinion that the sun had risen at the same point but had been rising at a narrower angle.
For this reason the four cardinal points of Homer superimposed on our rose of winds have the following positions: Since the direction of Notos is determined by the sun at the highest point, it corresponds to our South, and Boreas bend the direction opposite to Notos corresponds to our North. But Euros and Zephyros are at a point 30 degrees to the north of our East and West.
That this is the system of orientation of Homer can be readily proved by considering the way in which a more scientific system of orientation was introduced. The pseudo-Hippocratean writing On the Number Seven (middle of the fifth century B.C.) suggests that three different points of rising and setting of the sun can be considered: that at the Summer solstice, that at the Equinoxes, and that at the Winter solstice. From this there results a rose of winds of this sort with eight orientations:
Aristotle uses the same scheme but places the names of the winds in a different order. It is clear from his text that he intends to present a novel scheme in suggesting that the points of the horizon be placed on a circle. He repeats many times tha the opposite point to the point of rising of the Sun at the summer solstice is not the point of setting at the Summer solstice, but that a the Winter solstice, and that the same applies conversely to the point of rising of the sun at the Winter solstice. He argues in great detail that diameters have to be drawn and not chords. He constructs the following figure: