MAX-PLANCK-INSTITUT FÜR WISSENSCHAFTSGESCHICHTE Max Planck Institute for the History of Science 2013 PREPRINT 445 Horst Nowacki Archimedes and Ship Design Archimedes and Ship Design1 Horst Nowacki, Prof. em. of Ship Design Technical University of Berlin, Germany Max Planck Institute for the History of Science, Berlin Abstract In his written work Archimedes’ primary contributions to ship design were the Law of Buoyancy and the Criterion of Stability of a floating object. These laws form the foundations of the floatability and upright stability of ships. How did he create and justify this fundamental knowledge? Might he have applied it to contemporary issues in the ship design of his era? How was this knowledge passed down through the centuries and when and how was it applied to practical ship design decisions? What is the role of Archimedes’ fundamental insights in today’s ship design? This article will address these issues in its three sections on Archimedes’ own contributions, the history of his heritage in the maritime field and the continuing significance of his physical laws in modern ship design. Motto on Archimedes: All praise him, few read him, all admire him, few understand him. (A. Tacquet, 1612-1660). 1. Introduction Archimedes by many is regarded as the most eminent mathematician, mechanicist and engineer in antiquity. He is also famous for his practical application of scientific knowledge in engineering design. His knowledge has influenced ship design for many centuries, either directly or indirectly. Yet it took two millennia before his basic insights were applied quantitatively in practice at the design stage of ships. Why this long delay? It appears worthwhile to trace the history of this tedious knowledge transfer from antiquity to modernity. How were these elements of knowledge created and justified by Archimedes, how were they passed on as his heritage and how and when were they applied in ship design practice to this day? Ship safety and stability considerations play a dominant role in the ship design decision process. One needs fundamental knowledge, based on physical principles, to design a safe ship. This kind ofknowledge may be applied in two forms: • As intuitive, qualitative knowledge, corroborated by observation and experience, augmented by a rational understanding of the mechanisms of stability. This sort of cause and effect feeling guides in numerous tradeoff decisions on practical consequences of design measures. • As a quantitative knowledge, based on calculations, to predict the stability performance of a new design. Archimedes’ principles have been exerting a strong influence in both categories, at least among those properly initiated. But it was a long and arduous road from his first creation and brilliant justification of the basic concepts of hydrostatics, which lay buried in rare copies of scrolls in scientific libraries in antiquity, to their broader spreading after rediscovery during the Middle Ages and their wider circulation and gradual acceptance by printed media and at last to their practical application in design calculations. They have turned into routine design tools today. It is almost a miracle that we benefit today for the safety of our maritime designs 1This article was written for and presented at the International Conference on “Archimedes in the 21stCentury”, held at New York University, NYC on the 31stof May/ 1stof June, 2013, and organized by Prof. Chris Rorres. 1 from ideas conceived more than 2000 years ago, then almost forgotten, but luckily resurrected and brought to useful application only after about 1750. To recount this history and to understand how this happened the following questions will be addressed in this article: - How did Archimedes create the fundamental knowledge and justify the laws of hydrostatics which until today form the foundation for judging ship floatability and stability in ship design? - How might Archimedes himself have applied these insights to contemporary practical projects in ship designin his era? - On what circuitous routes did Archimedes’ insights arrive in modernity? - How and when were the laws of Archimedes, supplemented by other fundamental methods, apt to be applied quantitatively and numerically to practical ship design calculations? - What role do Archimedes’basic laws play today in modern ship design methodology? It is the purpose of this paper to trace the history of the knowledge in ship floatability, stability and design created by Archimedes and passed down through many centuries to its current relevance in modern ship design. 2. Archimedes 2.1 Precursors Human knowledge on the risks of seafaring is ancient. The first human experiences in oceangoing navigation by waterborne vehicles date back well into prehistoric times. There is indirect, though conclusive evidence of the first human settlers on the continent of then contiguous Australia/New Guinea arriving from mainland Southeast Asia between about 30000 to 40000 years ago by crossing a deep ocean water gap of about 100 miles using watercraft capable of carrying humans, animals and cargo (Diamond [1]). In the Mediterranean Sea traces of waterborne navigation date back to about 10000 B.C., notably in Egypt, Babylonia, later in Phoenicia, Crete, Cyprus and Greece. Early ocean voyages, some of them across considerable distances, required seaworthy ships, safe against all hazards of the sea. Details are documented in the literature (Kemp [2], Johnson, Nurminen [3]). Thus at the time of Archimedes during the 3rd c. B.C. Greek shipbuilding and ship design, as he knew it, had already reached an advanced level of construction technology and design complexity for ships to be used in trade, cargo transport and even warfare. Safer and ever larger ships had evolved during many centuries. The safety of ships was judged, as today, by their ability to survive the risks of sinking and capsizing, i.e., by their floatability and upright stability. These properties had to be assessed largely intuitively on the basis of experience, observation and comparison with similar designs. It was difficult to predict the performance of new designs prior to building them for lack of physical insights as well as analytical and numerical methods of design evaluation. The rational foundations for design decisions and for safe ship operations had not yet been laid. It was Archimedes, the eminent mathematician, mechanicist and engineer, who in a stroke of genius was able to combine these various viewpoints and to develop a rational theory of hydrostatics of floating objects, which was directlyapplicable to the issues of ship floatability and stability. Thus the physical principles of hydrostatics were then well understood although the practical application to ship design was long delayed, actually by about two millennia, before numerical calculations of these important elements of ship safety could be performed routinely by numerical methods at the design stage. We will discuss the reasons for this long delay in a later section. Nevertheless Archimedes and his contemporaries on the basis of his 2 hydrostatic principles were able to judge at least qualitatively the deeper physical processes involved in ship hydrostatics and thereby to judge the potential effectiveness of certain design measures. On the mathematical side and in the logic of rigorous proofs Archimedes had several predecessors, too, in the tradition of earlier Greek philosophy and mathematics. Table 1 gives an overview of some important precursors and contemporaries who influenced his work. It is important to single out Eudoxos [4], a pupil of Plato, who not only established rules and set standards for the rigor demanded in Greek proofs, but also worked out the Method of Exhaustion, an approximation scheme for evaluating integration hypotheses for curves and surfaces by successive refinement of inscribed or circumscribed polygons (or polyhedra). Archimedes made much use of this approximation technique, which was his tool for area, volume and centroid evaluation of simple geometric figures, while integral calculus was not yet conceived in antiquity. Archimedes was well trained in the contemporary methods of Greek mathematics, both by his early education received in Syracuse, reportedly also by his father, an astronomer, and by his contacts with and probable visit to Alexandria, an ancient center of excellence in mathematics in this era. Details on the knowledge thus available to him are found in Heath [5]. Table 1: Chronology of Precursors and Contemporaries of Archimedes Thales of Milet (624-544 B.C.) Pythagoras (580-496 B.C.) Demokritos (ca. 460-ca. 360 B.C.) Plato (427-347 B.C.) Eudoxos (410-356 B.C.) Aristotle (384-322 B.C.) Alexandria founded: 332 B.C. Euklid (325 B.C.-ca. 265 B.C.?) Mouseion in Alexandria: 286-47 B.C. Archimedes (ca. 287-212 B.C.) Eratosthenes (284-204 B.C.) Thus Archimedes was able to build his deductions on a solid tradition in Greek mathematics and logical rigor, but he also was very creative in developing his own mathematical tools whenhe needed new, original ones. 2.2 On Floating Bodies • Preparations Archimedes of Syracuse laid the foundations for the hydrostatics of floating objects in his famous¡treatise¡ﭛOn¡Floating¡Bodies¡(OFB)ﭜ¡(Ϸςϸὶ¡ὀϾόϼωέϊЀϊ)¡[6]. In this treatise he was the first single-handedly to establish the laws of equilibrium for a body at rest in a fluid, floating on top or submerged or even grounded, on a scientific basis by deduction from a few axioms or first principles. Thus although he never wrote about applications to ships he did develop 3 the physical foundation for judging the force and moment equilibrium of floating objects, including ships, i.e., their floatability (force equilibrium) and stability (moment equilibrium). The Principle of Archimedes, based on force equilibrium of buoyancy and gravity forces, holds for objects of any shape. The criterion of stabilitywas first pronounced by Archimedes for the special case of homogeneous solids of simple shape, a semi-sphere and a paraboloid of revolution. These results form the cornerstones of ship hydrostatics to this day. We are fortunate that many, though not all, of Archimedes’ treatises have survived from antiquity to this day, essentially all in copies, some in Latin translation, some in Arabic, and a few even in the Greek language of the original. Many were originally lost in late antiquity and were often luckily rediscovered much later, which will be addressed in a later section. But from those which are preserved we are able to reconstruct the train of thought that Archimedes took to arrive from the principles of geometry and engineering mechanics at his scientific foundation of hydrostatics. Table 2 gives an overview of the essential preserved treatises by Archimedes. We cannot precisely date the first appearance of these works, but there is sufficient evidence in their contents to suggest their sequence of publication. Compilations of Archimedes’ works exist in several classical and modern languages (Heath [5], Heiberg [6,7], Dijksterhuis [8], Van Eecke [9], Czwallina-Allenstein [10, 11] etc.). They are essentially in agreement on the chronology of appearance. Thus it is in essence undisputed that OFB was preceded by a few other fundamental treatises, which we will briefly address here. Table 2: Chronology of Archimedes’ Preserved Treatises Probable Item Title Sequence On the Sphere and Cylinder, Books I and II (5) 1 Measurement of a Circle (9) 2 On Conoids and Spheroids (7) 3 On Spirals (6) 4 On the Equilibrium of Planes, Books I and II (1) and (3) 5 The Sandreckoner (10) 6 Quadature of the Parabola (2) 7 On Floating Bodies, Books I and II (8) 8 Stomachion 9 The Method of Mechanical Theorems (4) 10 Book of Lemmas 11 The Cattle Problem 12 • The Law of the Lever In his treatise “On the Equilibrium of Planes”, Books I and II, Archimedes concerns himself with the “moment” equilibrium of objects on a lever system like a balance (Fig. 1). The 4 objects might be homogeneous solids or elements of thin planar areas of constant thickness and homogeneous gravity distribution, which can be regarded as solids so that the same lever laws can be applied in dealing with their equilibrium and centroids. The principle of the lever and especially of the balance was certainly known since prehistoric times, e.g., in ancient Babylonia, Egypt and China (see Sprague de Camp [12], Renn and Schemmel [13]). Archimedes presumably was the first to pronounce the physical law of the lever and to apply it to many mechanical and geometric systems. The Greeks did not know the concept and terminology of moments, so he spoke of “The Law of the Lever” for the same purpose. Fig.1: Lever System, Balance with Unequal Arms (from [10]) In this treatise Archimedes derives the following conclusions from the Law of the Lever: - The equilibrium of unequal weights on a balance of unequal arms (Fig. 1) so that the weights are inversely proportional to their lever arms (“moment equilibrium”). - How to lump two or more objects into a single compound object with a compound centroid so that the statical moment of the compound object remains the same as for the sum of the separate objects. - Removing, adding or shifting objects in a system and finding the new centroid. To summarize, in this treatise Archimedes pronounces the following principles, to which he can later resort for the stability of floating bodies: - The Law of the Lever (moment equilibrium). - Lumping objects into their centroids, thus forming resultants. - Finding compound centroids for a set of system components. - Removing, adding or shifting objects in a system and its effect on the centroid. • The Method of Exhaustion In evaluating the area of planar figures or the volume of geometric solids the ancient Greeks met with two major obstacles: - They had no concept of real numbers, let alone irrational or transcendental numbers, nor did they have a consistent system of units for measuring length, area and volume. They circumnavigated this difficulty by restricting themselves to finding only the ratio of a figure area or volume to that of a known figure, e.g., a square or a triangle whose size is known. In this context real numbers were expressed by ratios of integers. Irrational or transcendental numbers were approximated as such ratios. E.g., finding the area of a circle relative to a circumscribed square led to the famous “quadrature of the circle” type of problem. 5 - They had no method to evaluate areas and volumes of figures equivalent to modern calculus. This is why they resorted to approximation methods based on a finite number of successive subdivisions of the given figure by simple shapes whose area or volume are known. The Method of Exhaustion, as it was named much later in the 17th c., is such an approximation method based on successive refinement of the result by means of a polygonal approximant whose deviation from a given figure shall be made as small as desired after a sufficient, finite number of subdivision steps. The method is not equivalent to integral calculus since a limiting process to infinitesimal step size is not performed. But for geometrically well defined figures of simple shapes very accurate approximations can be obtained after a finite number of steps. To illustrate how Archimedes used this method let us refer to to his treatise “Quadrature of the Parabola”, §21-24, where he derives the area of a parabola segment AZBHC, compared to an inscribed triangle ABC whose apex is B such that the center line BD halves the parabola segment. Fig. 2: Parabola Segment AZBHC and Inscribed Triangles(from [10]) The Method of Exhaustion then proceeds as follows: Two additional triangles AZB and BHC are constructed between the parabola and the original triangle by drawing parallels to the center line BD through the quarter points E and K of the base line. For the parabola it can be shown that each of the new triangles has an area of (1/8) of triangle ABC, hence both together of (1/4) of the original triangle. In a process of successive refinement in each step new triangles are added between the polygon sides and the parabolic arc, always halving the intervals along the base line. It can be shown that for the parabola each new set of triangles adds (1/4) of the area of the preceding set of triangles. Thus in a process of continuing refinement the terms of triangle areas are added to a geometric progression whose quotient is (1/4). The sum of this progression is either known on arithmetical grounds, viz., for the parabola, or estimated after a finite number of steps when the truncation error appears small enough. For the parabola the end result is (4/3) times the area of the first triangle ABC. Such a result is first asserted inductively, then proven rigorously by reductio ad absurdum of any deviating results. Details of this proof can be found in Archimedes’ original treatise (cf., e.g., Heath [5] or Nowacki [14]). Archimedes, who seems to have adopted this method from Eudoxos, frequently used it for the evaluation of areas and volumes of simple shapes, i.e., for certain applications where modern analysis would use integral calculus. This explains why certain of his results were limited to 6 simple geometric shapes, e.g., for the hydrostatic stability of floating objects, and were not extended to objects of arbitrary shapes, e.g., ships. • The Method of Mechanical Theorems In certain cases Archimedes also had another fast and efficient method available to derive hypotheses for geometric results from mechanical analogies. This technique is described in his treatise “The Method of Mechanical Theorems”, which was long lost, but then was rediscovered first by J.L. Heiberg in 1906 in a Greek monastery, the Metochion, in Constantinople in an old 12th c. palimpsest. The Archimedes text had been rinsed off and a Greek prayerbook (euchologion) was written on the same vellum sheets. But Heiberg was able to decipher most of the original Archimedes text under a magnifying glass, to document, transcribe and translate it into German [15]. The interesting history of this palimpsest is described in more detail in Sections 3.2and 3.3. In this treatise Archimedes explains to Eratosthenes how he applies principles of mechanics in geometric reasoning to obtain inductive conclusions on geometric facts. Mechanical theorems are based on observation, hence inductively founded. Archimedes therefore uses these methods only to help propose hypotheses on geometrical facts. Archimedes does not regard these results as validly proven, but a strict, deductive, purely geometric proof must follow to confirm the conjecture. Some of these subsequent geometric proofs are preserved, others are lost. In his treatise “The Method…” Archimedes deals with the example of a solid paraboloid of revolution. He asserts: “The centroid of a right-handed conoid (here a paraboloid of revolution) cut off by a plane at right angles to its axis lies on the straight line which is the axis of the segment, and divides the said straight line in such a way that the portion of it adjacent to the vertex is double of the remaining portion”. The proof, based on the Law of the Lever and “The Method…”, is presented in detail in the treatise “The Method…” (Heath [5]), proposition 5, and is also fully explained by Nowacki [14]. This confirms that the centroid of the paraboloid segment is located at 1/3 of segment height above the base line or 2/3 below the summit. Thus Archimedes was able to demonstrate geometric facts without needing to resort to calculus. Archimedes later uses this result in the context of his study on hydrostatic stability of a paraboloid in OFB. • The Principle of Archimedes The fundamental law of hydrostatics for a body at rest within or on top of a homogeneous liquid is pronounced and justified by Archimedes in his treatise OFB, Book I. It is deduced here strictly by an experiment of thought, i.e., without any experimental observation or other empirical basis. It holds for a body of arbitrary shape. The liquid is assumed to be homogenous and such that any liquid particle is pressed downward vertically by all particles in the vertical line above it (OFB, Book I, § 1). We would call this a hydrostatic pressure distribution in modern terminology, but the Greeks did not know the concept of “pressure” in antiquity. In Book 1, §5 Archimedes asserts for a body in a liquid at rest: “A body submerges in a specifically heavier liquid to the extent that the volume of the liquid displaced by it weighs as much as the whole body”. In modern terminology this Principle of Archimedescan be stated as: 7 Π π V = , where Π = displacement = weight of body V = displaced volume π = specific weight of liquid V π = weight of displaced liquid volume = = buoyancy force Fig. 3: Proof of Archimedes’Principle (from [10]) Archimedes’ proof in OFB, Book 1, §5 is brilliantly elegant and brief (Fig. 3): • The surface of any liquid at rest is a spherical surface whose center point is the center of the earth (section ALMND). • The body EZTH be specifically lighter than the liquid, hence it floats in the surface. • We consider two neighboring equal sectors of the sphere, bounded by the surfaces LM and MN. The first sector contains the floating body whose submerged part is BHTC. The second sector instead has an equal volume RYCS filled with the liquid. • The liquid is at rest. Thus the surfaces between XO and OP experience identical pressing loads. Thus the weights of the volumes on top of these surfaces are equal. But the weight of the liquid in the first sector, aside of the space BHTC, equals the liquid weight in the second sector, aside of the space RYCS. Therefore clearly the weight of the body EZTH equals the weight of the liquid volume RYCS. It follows that the liquid volume displaced by the body weighs as much as the whole body. Note that this concise proof of the Principle of Archimedes is strictly deductive and is based entirely on an experiment of thought and very few axiomatic premises. It holds for arbitrary body shapes in an arbitrary type of liquid. It was derived for liquids and bodies at rest without knowledge of the pressure distribution anywhere. • The Eureka Legend The famous Eureka legend about Archimedes’ discovery of a hydrostatic law in the bathtub, which goes back to Vitruvius ([16], Book IX.3, published around the birth of Christ) is often misinterpreted and has led to confusion regarding this discovery. According to Vitruvius Archimedes was challenged by King Hieron to determine whether a wreath, made for the king by a goldsmith for a votive offering to the gods, was of pure gold or fraudulently made of gold mixed with silver. Archimedes is said to have observed how the water displaced by his body in a brimful bathtub was a measure of the displaced volume, which he regarded as a 8