Energy and Society: Chapter 3: Inorganic Energy Sources: Wind and Water


Man has long sought to harness wind and water power efficiently. He has been hampered in that effort by the nature of these forces, which makes the converters that use them comparatively costly. In many cases the kind of energy they delivered was not very useful in his culture.

The river as transport

The flowing stream was undoubtedly used as a means of transportation long before written records could testify to this use. The more complicated problems of delivering the energy of the stream at some fixed point were not solved until much later. The water wheel came into use where food raising made reciprocating or rotary motion a valuable adjunct to energy from the older converters.

Both the flowing stream, as a carrier, and the water wheel expanded the limits of the culture of the men using them. As an energy source, the stream differs greatly from plants and animals. Within its ribbon like field, energy is constantly available, but just beyond those limits the stream provides no energy. It thus sets a spatially defined pattern for those who use it. It can carry great loads cheaply but only within well-defined areas. Most of the great streams provide energy for only a small part of the inhabitable area adjoining them. So where great areas were engaged in food raising, only a few workers could deliver their surplus to the stream to be borne away by it without consuming a fairly large part of that surplus in the costs of transportation to the stream’s edge. In other words, the streams in areas predominantly agricultural provide an energy advantage for only a thin fringe of communities located along river banks. Beyond the points where it is advantageous to use the stream the centripetal effect which naturally arises from the use of plants becomes manifest, that is, there is the tendency, already noted, to limit the size of settlements by reason of the energy costs entailed in going to and coming from the fields. The same facts tended to make settlements almost completely self-subsistent.

Moreover, the stream provides surplus energy only in one direction, downstream. It moves the product of surplus energy produced under the sun-plant-animal system away from its original site. If reciprocity or some kind of equilibrium between downstream and upstream areas is to be established, energy must be expended in returning to upstream users goods in exchange for those received. If this round-trip relationship were to represent the exchange of equal masses of goods, the stream would cease to be a source of surplus, for at least as much energy must be expended in getting goods upstream as was gained in floating an equal heavy cargo downstream. Under such conditions the stream would become a highway rather than an energy source.

These physical facts limit the kinds of culture which can make effective use of the energy of the stream in transportation. They also limit the effect which energy can have when put into use.

As I pointed out earlier, where fertile soil exists in sufficient quantity in proportion to the population which is used in tilling it, it is usually possible to recover considerably more energy in the form of the plant product than goes into producing the plant. In many areas of the world where this situation obtains men tend to multiply until the proportion of land to population is such that the energy secured from food raising is equal only to that which is required to sustain the local population. This is not a cause of extinction for the self-subsistent community, but it destroys all basis for extramural trade. The flowing stream is thus most useful when some relationship serves at once to limit population to the point of maximum energy surplus and to concentrate control over that surplus so that the largest possible portion of it can be used to produce goods for trade. Slavery and serfdom offered this combination of circumstances. They usually resulted in limiting population, for the serf was, in return for certain duties, supplied with a fixed amount of goods, regardless of the amount of mouths he was attempting to feed, and the slave was provided with only sufficient goods to supply the number of children required to serve the master. Slavery and serfdom likewise frequently resulted in a concentration of control over surplus, which was sometimes utilized in trade for luxuries. It was in fact as an adjunct to such systems that the stream made its original contribution to civilization. It is frequently true that such exploitative systems run contrary to values which recur in family and community life, so that if a breakdown in power occurs the upstream communities tend to restore local autonomy in the control of surpluses and reduce exports in favor of local use of those surpluses. River trade declines and with it surplus energy from the stream.

Only if some kind of reciprocity exists between those who accumulate at a downstream site the product of upstream cultivation and those who make such surplus available can the relationship be continued. The values of those upstream who control surpluses must be complementary to the values of those with whom they deal downstream. Such trade characteristically takes the form of the receipt, from upstream sites, of raw materials, particularly minerals, forest products, food, and fiber, which have considerable bulk and mass. In return for the product of artisanship, or of special skill or knowledge, which weighs less, is delivered upstream. This arrangement effects transport upstream at considerably less energy cost than is involved in the downstream movement of goods. It also frequently encourages upstream landlords to continue their exploitive position. Thus the downstream site is likely to produce luxury goods for a small ruling class upstream in return for goods that might have been essential to the survival of many upstream villagers had they not been taken for the benefit of the trader. The historic antagonism between downstream towns and those located in the hinterland thus has a very natural origin.

Early civilization developed along the Nile, the Tigris, and the Euphrates. Cities also developed on the rivers of India and China as the spread of cultivation there brought about sufficient increase in surplus energy. But before the rise of agriculture there was no similar early development on the great rivers of North and South America, Europe, South and Central Africa. Early transportation by stream is thus seen to have served primarily as a limited adjunct to food-raising societies with the cultural means to concentrate available surplus. It must be remembered, however, that the social conditions required for the maintenance of agriculture also set limits to the kinds of civilization which could grow up around the use of the stream as a source of energy.

The river as a source of mechanical energy

The appropriation of the energy of the flowing stream through its delivery at a fixed site is a much more complicated matter that the use of water for transportation. The first of the limits on the generation of mechanical energy form the stream is to be found in the potential energy stored in the water. As was pointed out earlier, a horsepower is equivalent to the delivery of 550 foot-pounds per second. This amounts to 1,980,000 foot-pounds per horsepower-hour. In other words, if the converter used is 100 percent efficient in turning the potential energy of the water into mechanical energy, a ton of water must fall about ¼ mile to deliver the equivalent of 1 kilowatt-hour. Now the great areas in which food raisers create their largest surpluses are old river deltas, areas of very low relief. The Mississippi, for example, falls only 322 feet between Cairo, Illinois, where it joins the Ohio, and the Gulf of Mexico, more than 600 miles away. It falls only about 800 feet between Minneapolis and the Gulf, a distance of over 1,000 miles. Great rivers are so huge in volume that their potential energy is enormous even with so slight a fall. But means of handling such quantities of water as would permit the delivery of any great amount of mechanical energy with only a slight fall are even today prohibitively expensive. Moreover, the flow in any of the major rivers varies greatly during the year, and any works set up to capture their energy must allow for variation in the depth of the stream. Dams and locks which will withstand the force of great depths of moving water are generally considered to be more costly than is justified by the returns, as compared with other alternatives. Consequently any use of the water wheel on large rives is likely to be limited to temporary structures or floating mounts. But these create new problems involved in delivering mechanical energy to the river banks. On small streams, where the flow is more constant and the flow is more constant and the descent of the water precipitous, water wheels have often provided an important adjunct to the power of limited areas.

Limitations of water wheel and windmill

But here we encounter another characteristic of both water wheels and windmills. This is the character of the field of the converter. To transform the power of moving wind or water into a localized force, the weight or force of the water or wind, acting on a vane or cup, is converted into angular motion about the axle of the wheel. Thence, it must be transmitted as reciprocating or rotary motion. The area over which it can be delivered efficiently is determined by the efficiency of moving cams, shafts, belts, or gears. Such means of transmission quickly waste, in the form of heat generated through friction, much of the energy which they are intended to transmit. So the gradient in the field of a converter using mechanical transmission is very steep.

Now in food-raising areas there are few tasks in which rotary or reciprocating motion with a narrowly limited field are involved. The bulk of the energy required, at the time when energy is most scarce, is used in plowing or otherwise preparing the seedbed or in harvesting. In this task energy must be used over a large area, and thus the water wheel or windmill is ill-fitted to carry it out. It is true that at some other times of the year reciprocating or rotary motion can be used, as in separating grain from chaff and particularly in milling it, or in spinning and weaving fiber. But there are often many hands which are idle except when preparing the seedbed, harvesting, and perhaps irrigating. These hands are well fitted to carry out exactly the kind of task for which the windmill or water wheel is also fitted. We thus come to the same kind of competition that eliminated the use of animals for draft in large parts of the world. If energy is diverted into the building and maintenance of the wheel, but the wheel adds nothing to the total food supply, it provides no useful energy in addition to that which is otherwise available in the form of unemployed or underemployed persons. In such cases diversion of energy to create converters merely reduces the supply of energy which could otherwise be used to increase directly the supply of goods or services sought as ends.

For example, a woman may expend less energy in grinding meal in a hand mill or metate or in pounding it out with a pestle in a mortar than she would expend carrying the grain to the mill and paying for the service rendered there. It may take more energy to transport unthreshed grain to the power separator, pay for the service, and return the grain than to flail it out and allow the wind to separate it, or to use a draft animal to stamp it out on a threshing floor. Thus only on extremely fertile lands, producing large crops in a very small area, could a fixed-site threshing or grinding machine be supplied with work enough to keep it going much of the year. Transportation costs frequently preclude the economic operation of the mill elsewhere. Landlords sometimes used water-and-wind-driven mills to increase their own supply of energy by forcing their tenants to use the mills, taking part of the grain or meal in return, but frequently this merely increased the “unemployed” time of the peasant and left him with less food to consume than would otherwise have been available to him.

Other operations using rotary and reciprocal motion common to agricultural communities are spinning and weaving. These are laborious, but again they are among the means by which the hands needed at plowing or harvest time are kept busy at other times of the year. They enable those who cannot labor in the field to establish a legitimate claim to a share of the product of the field. Except in hilly or mountainous areas the commonly used fibers are produced on land which can be used alternatively to raise food and fiber. Thus fiber may be locally produced if it can so be obtained at a lower cost than the value of the food which could be produced on the same land. A self-sufficient agricultural community may elect to establish reciprocal exchange with another area in which food can be produced only at cost much high than that of producing fiber. To mountain people, for example, the production of fiber such as wool may provide a means to secure food or other energy sources in amounts greater than could be directly secured from their land. But to the agricultural community trade is energy wise a means of gain only if the food to be exchanged plus the energy involved in its transport and the transport of the fiber cost less to produce than does the fiber to be secured in trade.

Given the presence of frequently idle hands — which is characteristic of many food-raising areas — the energy costs of spinning and weaving are not as significant as the net loss in energy that would result if food were to be exchanged for fiber, since the fiber can be secured with less energy at home. In certain areas, such as Iran, there are highlands which will support sheep and goats in greater numbers than human beings, and these areas adjoin valleys in which food can be produced to exchange for cotton and other fiber with less land than would be required to raise the cotton or other fiber. Distances are short enough to make transportation costs low, and as consequence fairly large groups subsist in these highlands primarily by producing fiber for exchange. Most generally, however, fiber can be produced as a side line of the self-contained agricultural areas. Towns in which there is more work to be done in producing textiles than can be supplied by the idle hands are rare. It was only when transportation costs were radically reduced that converters to be used specifically for spinning and weaving became economical in terms of the values characteristic of the self-contained village or manor. The use of the water wheel for this purpose was thus very limited.

Special characteristics of the windmill

The windmill met with many of the same obstacles. While the wind does not fix the location of the windmill so exactly as water does that of the water wheel, the windmill, like the water wheel, delivers its energy in a limited field characterized by a steep gradient. So again the only economical use of its energy is in the form of reciprocating or rotary motion. There are also, however, differences between the use of the windmill and that of the water wheel which arise from the character of the source of the energy. Wind is generated by changes in the heat reaching various parts of the earth and moves in somewhat predictable paths, but these exhibit none of the sharp boundaries which characterize streams. And though in some areas where basic crops can be grown the wind is fairly predictable, it is hardly ever constant. The average velocity of the wind in a place tells little about the actual power potential there. Wind blowing 12 hours at 60 miles an hour and falling to a dead calm during the next 12 hours has averaged 30 miles an hour, but the total energy delivered is not the same as it is in a place in which it has blown steadily at 30 miles throughout the day and the night.

The pattern of the winds, the periods of time both seasonal and other during which they blow, has much to do with the power that can be made available. Windmills are most useful, then, where the winds blow at such velocities and at such times that they neither allow the converters to stand idle nor submit them to damaging strain. Such sites are rare. In a few areas, such as Holland, where their use could be depended upon to keep water from inundating low fields, or in the American Far West, where sufficient water to supply culinary needs and those of livestock could be stored to last until the wind blew again, the windmill proved to be a valuable adjunct. In the West Indies trade winds provided the power to mill great quantities of sugar cane which trade permitted to be grown there. With but a few such exceptions, however, the windmill altered none of the established ways of exploiting the energy of plants and animals.

So long as only mechanical means existed for the transmission of the power of the water wheel and the windmill, they remained incapable of producing any sweeping changes in society. For the power they deliver today the world had to await the development of electrical generating and transmitting equipment, which originally depended on the use of other converters such as the steam engine. Accordingly, this use of wind and water power will be discussed later.

The sea and the ship

When the flowing stream reaches sea level, the potential energy it carries is exhausted. Except in the ocean currents, oceans, seas, and lakes provide as such no energy to those who use them, though they serve as cheap highways in many cases. Thus boats and ships must be propelled by other sources. Those propelled by men or animals are subjected to the limits of which we have hitherto spoken. The wind-driven vessel is another story. The advent of the sailing ship was potentially capable of working a revolution among all those societies located on navigable waters. Unlike the use of animals as a source of mechanical energy, the use of the wind does not diminish the energy available to man in the form of food. The energy costs of operating a ship are only those of building, maintaining, and manning it. The surplus energy derived from the sails is potentially enormous as compared with the cost of producing the sail and hoisting it. Thus, for the first time in man’s history, men, using the sailing ship, came into control of very large amounts of power largely independent of plant life or of the number of persons in the population using it. Sailing vessels are confined to waters in which they can maneuver and in which the winds are such as to propel them. Within these limits upon the trackless seas the sailing ship delivered energy at a rate previously unimagined.

The gradient of the sailing ship is more gentle than that of any of the converters previously discussed. For example, the amount of food a man or an animal can carry on his back or pull behind him in a wheeled vehicle is tiny as contrasted with the amount of food a sailboat managed by one man can carry. In addition, the distance a sailing ship can cover in a day, under favorable conditions, may be much greater than the distance a man walking or riding an animal can cover in the same time. Moreover, and this became very important historically, the field of the sailing ship merges with that of the flowing stream at its mouth, and energy derived from one of these sources can be used to supplement that derived from the other in maintaining or creating desired relations between downstream cities and the interior. Before we examine the new sets of social relationships to which the sailing ship contributed, it may be wise to inquire into the physical characteristics of the sailing ship as a converter.

How much energy does the sailing ship deliver?

No ratings in modern terms of the various types of sailing ships exist. At the time they were built, no calculation of the horsepower required to move them was made:  a ship that had a good record, that provided adequately for a cargo, and that had low operating expenses was simply copied when another ship that was likely to sail in the same waters and to carry a similar cargo was needed. Modifications to achieve other aims or correct weaknesses that become apparent in operation were made on a trial-and-error basis.

Today ship architects usually design and test models, determining which shapes are desirable on the basis of the energy required to pull the various models. They can then calculate the power which will be required to drive the full-sized ship at the speed sought. In the construction of smaller vessels, particularly, where allowing extra power to compensate for possible error involves less cost than does the constructing and testing of models, set formulas are used, but at best such formulas are somewhat inaccurate. It is thus difficult to calculate the power necessary to drive ships which have long since been sunk or dismantled. But for the most part there are exact figures on the size of those ships and the speed of which they are capable. The horsepower required to drive them can be roughly calculated by direct comparison with that used today to drive ships of similar size at comparable speeds. This method leaves much to be desired as to accuracy, but in view of the enormous difference between the surplus produced by the ship and that produced by earlier converters, whatever error creeps in is probably not great enough to radically affect our conclusions.

The efficient size of the ship at any time or place is determined by such factors as available materials, existing shipbuilding technology, nature of the cargo, means of loading and unloading, the ability of the crew to put on and take off sail (taking into account the winds to be encountered), the purpose for which the vessel is be used, etc. At various times one or another of these factors historically constituted the ruling limit. On the other hand the size of the crew was related not only to the size of the ship but also to such conditions as whether she was to be expected at times to depend upon oars, and whether she must to be a fighting ship. During long periods the right of transit was likely to be challenged by pirates, privateers, and “legitimate” operations of hostile navies. In these circumstances the crew could not be limited to what was technologically efficient, but was enlarged for purposed of defense. The surplus energy produced was correspondingly diminished. The Egyptians were able to design, build, and sail ships up to 150 tons.[1] There is little direct evidence of the speed of which they were capable, but from some indirect evidence it appears that it was more than 8 knots. A ship of this speed and displacement today requires engines that deliver about 80 horsepower. They carried a crew of about 40 men, so since each man consumed in the form of food the equivalent of about 1 horsepower-hour per day, the surplus developed at a maximum was about 47 horsepower-hours per day per crew member when the ship operated round the clock. This considerably in excess of anything earlier developed upon land. The Egyptians used the ship chiefly to gather such luxuries as myrrh form Somaliland. Later, the Phoenicians developed the art and techniques of shipbuilding and navigation further, but their ships, though somewhat different form those of the Egyptians, were hardly larger.

The Greeks developed the ship considerably. However, they were never able to secure such control upon the sea that they could reduce the crew to the point technologically most efficient. Their sailing ships, therefore, were also powered by oarsmen to increase their battle strength. The Greek trireme was manned by as many as 170 oarsmen on a boat 150 feet long and 16 feet wide, and her sail was only a supplementary source of power.

The Romans, after the defeat of the Carthaginians and the suppression of Greek piracy, were able to build true merchant ships up to 250 gross tons with crews reduced to the size necessary to handle the sails. At the speed of 8 knots these ships must have generated about 100 to 120 horsepower. With a crew of 10 to 12 they produced a maximum of 10 horsepower per crew man. Running 24 hours a day, this ship could thus generate about 240 horsepower-hours or 240 man-days per man per day. This maximum, or course, was rarely if ever realized. Contemporary Norsemen developed a ship of about 30 tons, 80 feet in length, which carried a crew of 90 men and made 10 knots. For a short period an oarsman can produce about 1 horsepower, so under stress the oarsmen could develop about 90 horsepower. But the sails, which could be operated round the clock, developed only about 30 horsepower. Hence, the surplus available to the Norsemen was quite small in comparison with that of the Romans. A vessel duplicating one of the early Norse ships was sailed across the Atlantic in 1893 with a crew of 12. This demonstrates that the Norse ship was not stripped to her minimum crew but, since it was chiefly a fighting vessel, was made to carry all the men she could, together with their arms and supplies, including booty.

Extensive trade was not characteristic of the use of the vessel outside the Mediterranean until medieval times, though the Phoenicians made such expeditions as those into the Baltic for amber during Rome’s era of greatness. In the 1400’s, increase in the surpluses available for trade, together with new navigational techniques and the invention of gunpowder, which permitted domination of coastal areas from the sea, led to the development of ships of 300 to 500 tons. The ships at 10 knots developed from 150 to 250 horsepower.

The sailing ship: revolution in energy source

The great transformation which came during the seventeenth and eighteenth centuries with Dutch and British exploitation of the sea was made with Indiamen of about a thousand tons. The rapidity with which the increase in the size of the ship took place can be seen from the fact that the flagships of both Columbus and Drake, a century earlier, were only 100 tons. Indiamen at 10 or 11 knots developed 500 to 750 horsepower with a crew of about 80 hands.

The climatic development of sail came in the early nineteenth century. Its achievements brought a surplus greater than that of any other converter in any preceding age. From the point of view of the total energy generated, if not of the surplus per man-day, the clipper ships were paramount. The record run of 436 miles in 23½ hours for an average of 18.55 knots made by the Lightning, a ship of 2,084 gross tons carrying a cargo of 2,000 tons, is respectable even today. The Sovereign of the Seas, a ship of 2,421 tons, is said to have once made 411 miles in a day, and single-mile runs at 22 knots or better have been authenticated for several of the clippers. This compares with the 23 knots of which the Empress of Japan, fastest ship in the Pacific in 1942 was capable. Other, larger, sailing ships were built, such as the Great Republic, 4,555 tons, the Maria Rickmers, 3,822 tons, and the Preussen, 5,081 tons, which averaged 16 knots for 24 hours, but all these ships made use of steam winches for hoisting and trimming sail and are thus excluded from our consideration of the surpluses generated by wind power combined only with manpower. The Sovereign of the Seas, which used only men to hand the sails, carried a crew of 105 hands and was able to make a top speed of about 17 knots. Since this was of course possible only with favoring wind, a steamship of somewhat lower-rated top speed under similar favorable circumstances would be able to equal it. Now modern vessels of about 2,400 tons, rated at about 14 knots, deliver 1,200 to 1,400 horsepower. Using the higher figure, we estimate that the Sovereign must have produced about 12 horsepower per crew man, or about 287 man-days surplus per man per day.

The greater effort required to attain the higher speeds did not prove to be justifiable except under unusual circumstances such as those surrounding the California gold rush or in the first voyages of the season, when such races as the Grain Races and the Tea Races were staged. The extra sails by which the clippers gained their speed, at the cost of a great increase in necessary crew, were discarded. The clippers were replaced by slower ships delivering a larger per capita, though a smaller total, surplus. The most efficient sailing ships were thus able to produce a maximum of 200 to 250 times the human energy required to operate them. The ideal limit was, of course, never realized for any very long period, owing to failure of winds, time in port, and other factors. Out of this operating surplus must come the amortization of capital costs, the development of necessary port facilities, etc. Nevertheless, the sailing ship was a source of surplus energy far in excess of that produced by any other known converter until well into the nineteenth century.

To repeat, the sailing ship produced large surpluses of energy and generated a field enormous in extent. Historically it was only the first of a number of such converters, all of which we shall call, in contrast to their organic prototypes, high-energy converters. Since it is obvious that sail offered tremendous advantages, it might be expected that every people living on the seacoast would have adopted and exploited it fully. Actually, very few areas did bring sail to any very high level of usage. This fact calls for further analysis.


  1. ^ Tonnage figures used here represent the carrying capacity of the vessel.  This is figured out by assigning 1 ton for each 100 cubic feet of usable space in the vessel.

This is a chapter from Energy and Society: The Relationship Between Energy, Social Change, and Economic Development (e-book).
Previous: Chapter 2: Organic Energy and the Low-Energy Society  |  Table of Contents  |  Next: Chapter 4: Sail and Trade



Cottrell, F. (2009). Energy and Society: Chapter 3: Inorganic Energy Sources: Wind and Water. Retrieved from


To add a comment, please Log In.