Tornadoes and waterspouts have long fascinated humankind through their presence in myths and popular beliefs and originally were believed to have supernatural causes. The first theories explaining weather phenomena as having natural causes were proposed by ancient Greek natural philosophers. Aristotle was one of the first natural philosophers to speculate about the formation of tornadoes and waterspouts in Meteorologica (circa 340 BCE). Aristotle believed that tornadoes and waterspouts were associated with the wind trapped inside the cloud and moving in a circular motion. When the wind escapes the cloud, its descending motion carries the cloud with it, leading to the formation of a typhon (i.e., tornado or waterspout). His theories were adopted and further nuanced by other Greek philosophers such as Theophrastus and Epicurus. Aristotle’s ideas also influenced Roman philosophers such as Lucretius, Seneca, and Pliny the Elder, who further developed his ideas and also added their own speculations (e.g., tornadoes do not need a parent cloud). Almost ignored, Meteorologica was translated into Latin in the twelfth century, initially from an Arabic version, leading to much greater influence over the next centuries and into the Renaissance. In the seventeenth century, the first book-length studies on tornadoes and waterspouts were published in Italy and France, marking the beginning of theoretical and observational studies on these phenomena in Europe. Even if speculations about tornadoes and waterspouts proposed by Greek and Roman authors were cited after the nineteenth century only as historical pieces, core ideas of modern theories explaining these vortices can be traced back to this early literature.
In recent years, there has been a revival in the study of European tornadoes and their impact (e.g., Antonescu et al. 2017; Taszarek and Gromadzki 2017; Shikhov and Chernokulsky 2018; Brázdil et al. 2019). This revival followed a period starting at the end of the Second World War that was characterized by a relative lack of interest on this subject from European researchers and meteorologists. Despite being an important and relevant topic of research, there has been no history of tornadoes in Europe, except for a few partial attempts (e.g., Peltier 1840; Wegener 1917; Hellmann 1917; Antonescu 2017). In this article, these recent efforts are extended back in time by examining the history of tornadoes in Europe, not from a climatological perspective (e.g., Groenemeijer and Kühne 2014; Antonescu et al. 2017), but from a history-of-ideas perspective. Thus, we focus on early theories1 of tornado formation as proposed by Greek and Roman natural philosophers.2 Our aim is not only to document these theories, but to understand their evolution and influence up until the seventeenth century when the first detailed observational studies of tornadoes and waterspouts were published in Europe.
This article is structured as follows. Section 2 presents the main ideas about the formation of weather phenomena as they appear in mythology and in the early natural theories. The theories on tornadoes and waterspouts proposed by Aristotle and other Greek philosophers are presented in section 3. Section 4 analyzes how the hypotheses about tornadoes and waterspouts proposed by Greek natural philosophers were adopted and developed by Roman poets and philosophers. Section 5 summarizes this article.
2. From mythology to natural theories
Weather phenomena have long been objects of fascination, curiosity, and threat for humankind. Ancient farmers and hunters were strongly dependent on weather conditions, and the ability to anticipate future weather conditions based on signs within nature would benefit their existence. Before discussing how two specific types of weather phenomena (i.e., tornadoes and waterspouts) were perceived in the ancient world, we need to understand the intellectual environment that led to theories explaining their formation. Thus, we need to look at how weather phenomena appeared in cultural traditions (e.g., mythology) and also at the first hypotheses about the formation of weather phenomena in general.
In the mythological works of ancient authors, meteorological phenomena (e.g., lightning, hail, tornadoes, waterspouts) were the result of divine intervention (Gregory 2006). Both Homer and Hesiod included numerous references to weather in their poems that were later quoted by other authors writing about meteorology (Taub 2003). In the Iliad [circa (c.) 750 BCE] and Odyssey (c. 725 BCE), Homer describes Zeus as being able to produce rain, thunder and lightning, and storms (Taub 2003, p. 5). Poseidon, the god of the sea, is also able to control the weather.
So he spoke, and he gathered the storm clouds, stirred up the seaway, wielding his trident in both of his hands; and of all of the winds he roused up all of the blustering gusts, and the sea and the mainland covered and hid in the clouds, while night rushed down from the heaven (Homer 2002, Odyssey 5, 291–294).
In Theogony (c. 700 BCE), Hesiod described the battle between Zeus and Typhon as a serpentine giant associated with a destructive windstorm (Fig. 1).
As the lord rushed forward, great Olympus trembled under his immortal feet, and the Earth groaned in response. The violet-dark sea was enveloped by a conflagration from both of them—of thunder and lightning, and fire from that monster of typhoons and winds, and the blazing thunder-bolt (Hesiod 2006, Theogony, 841–846).
The ancient Greeks did not just attribute the meteorological phenomena to gods (i.e., nonnatural explanations), but they were also probably the first to make regular meteorological observations and to propose theories (i.e., natural explanations) about the formation of weather phenomena (Hellmann 1908). Meteorological phenomena were ripe for hypotheses by natural philosophers such as those from the Milesian school. This school was founded around 650 BCE in Miletus, an ancient Greek city on the western coast of Anatolia, today in Turkey. Thales of Miletus (c. 624–c. 546 BCE), a Greek philosopher, mathematician, and astronomer, associated weather with the movement of the stars and planets (Frisinger 1971). Two of his followers, Anaximander (c. 610–c. 546 BCE) and Anaximenes (c. 585–c. 528 BCE), also tried to explain meteorological phenomena using natural causes. Anaximander was the first to give a definition of the wind as the flowing of air that happens when “the finest vapors, separated from the air, propelled when massed together” (Hippolytus 2016, book I, 7). He also speculated that thunder was the result of the air smashing against clouds and that lightning resulted from the violent separation of clouds by the wind (Frisinger 1977). It seems that centuries later these natural explanations were still considered contrary to the general view that weather phenomena were associated with gods, as shown by this dialogue from the comedy The Clouds. The dialogue, written by Aristophanes (c. 446–c. 386 BCE) in 423 BCE, is between Strepsiades, a countryman living in Athens, and the rationalist philosopher Socrates.
Strepsiades: […] But tell me who makes the thunder then, the sort that gives me the jitters.
Socrates: It’s these who thunder by rolling round.
Strepsiades: But how, audacious thinker?
Socrates: It’s when they’re soaked to the limit with water and compelled to move about while sagging low all teeming with rain, and then in this heavy state they collide with one another and make those sounds of cracking and rumbling.
Strepsiades: But who is it then who compels them to move, well isn’t it Zeus himself?
Socrates: Not at all! It’s the swirl of the atmosphere.
Strepsiades: The swirl? It was lost on me that Zeus just doesn’t exist but instead it’s swirl that rules the world. But you haven’t yet taught me exactly how the rumble and thunder occurs.
Socrates: Are you deaf? I told you it’s when the clouds are brimming full of water then bang into each other and rumble because they’re so compressed.
(Aristophanes 2015, The Clouds, 373–285).
At the beginning of the fourth century BCE, meteorology (i.e., “the study of things high in the air”) was in a period of decline. Wilson (2013) indicated that the philosophers from the Milesian school made further research in meteorology unnecessary by simply rejecting the nonnatural explanations as a cause of weather phenomena. Their speculations were later just accepted by other thinkers who considered that all the explanations for the formation of weather phenomena were provided and thus the Milesians focused their speculations on other topics rather than meteorology (Wilson 2013, p. 19).
3. Aristotle’s Meteorologica and the first theories on tornado and waterspout formation
Meteorology was reborn around 340 BCE because of Aristotle’s Meteorologica. Meteorologica was the first “systematic discussion of meteorology” (Zinszer 1944). Aristotle (384–322 BCE), the founder of the Peripatetic school (founded around 335 BCE) (Wilson 2013), reviewed the main theories existing at that time about weather phenomena and proposed new theories about rain, clouds, mist (book I, chapter 9); dew, hoar-frost (book I, chapter 10); rain, snow, hail in relation to hoar-frost (book I, chapter 11); hail (book I, chapter 12); winds (book II, chapters 3–6); and thunder and lightning (book II, chapter 9).
Aristotle’s theories on “hurricanes, typhoons, firewinds and thunderbolts” are described in book III (chapter 1) of Meteorologica. In building these theories, Aristotle did not discuss any actual events. To our knowledge, there are no surviving descriptions of tornadoes and waterspouts that occurred in ancient Greece. The earliest descriptions of tornadoes and waterspouts in Greece are from the eighteenth century (Matsangouras et al. 2014). The concept of tornadoes or waterspouts was introduced by Aristotle without any reference to previous theories or refutation of previous theories, in contrast to his theory about wind in which Aristotle refuted Anaximander’s ideas. Thus, Aristotle was likely the first natural philosopher to discuss the concept of tornadoes or waterspouts and to speculate about their formation.
To understand the theories about tornadoes and waterspouts proposed by Aristotle, we need to introduce some of the concepts that he employed to explain the formation of weather phenomena in general. Then, we will discuss the terminology used by Aristotle and subsequent ancient thinkers who wrote about tornadoes and waterspouts.
Aristotle described two types of processes that he called “exhalations” within the terrestrial region, the region below the orbit of the moon that is composed of four elements—earth, water, air, and fire—arranged in concentric spheres with Earth at the center (Frisinger 1972). According to Aristotle, these elements are in a continuous process of transformation from one to another. Thus, when the heat from the sun reaches the surface of Earth, it mixes with the cold and moist water to form the moist and cool exhalation (noninflammable, “essentially like air”; Frisinger 1972), which is associated with clouds and rain. When the heat from the sun interacts with the cold and dry Earth, the result is the dry and warm exhalation (inflammable, “essentially like fire”; Frisinger 1972), which is associated with wind, thunder, and lightning. A more detailed analysis of the concept of exhalations can be found, for example, in Frisinger (1965), Frisinger (1972), Taub (2003), and Wilson (2013). The moist and cool exhalation sinks, whereas the dry and warm exhalation rises. Both exhalations coexist, and when Aristotle was referring to one of them, he was referring to the one that is dominant for a certain type of weather phenomena. For example, wind is produced by the dry and warm exhalation as a result of the sun drying and heating Earth. As the dry and warm exhalation rises, only the vertical component of the wind is generated. The horizontal component of the wind is generated by the interaction between the dry and warm exhalation and the movement of the celestial spheres (i.e., a fundamental concept of the early cosmological models in which the rotating spheres made of a single matter called “aether” are responsible for the apparent movement of the stars and planets in the sky). Aristotle also indicated that the dry exhalation causes thunder and lightning when it is produced in small quantities, but when “there is an abundant and constant flow of exhalation,” εχνεφíαζ (ecnephias) is produced (Aristotle 1952, Meteorologica, 370b4–18).
In the Greek–English Lexicon (Liddell and Scott 1996), εχνεφíαζ is translated as “a hurricane, caused by clouds meeting and bursting.” The term “hurricane” was used in official meteorological reports, before the beginning of the twentieth century, for windstorms, and also to describe tornadoes and waterspouts [see, e.g., the discussion concerning the use of hurricane in tornado reports for Romania in the nineteenth century from Antonescu and Bell (2015)]. Thus, given Aristotle’s description of εχνεφíαζ, we argue that the term can be translated as a generic windstorm, which is the definition used in this article. Aristotle also made a distinction between windstorm and typhon (),3 both being associated with wind descending from the cloud, but the cloud trails behind the wind in the case of the typhon (Aristotle 1952, Meteorologica, 370b28–31, and the note by H. D. P. Lee on 234–235). Although the modern definition of the typhon (or typhoon) is a severe tropical cyclone in the western North Pacific Ocean (American Meteorological Society 2019b), Aristotle likely was referring to a tornado or a waterspout, which is the definition used in the remainder of this article.
According to Aristotle, a windstorm occurs as the result of the deflection of the wind within the surrounding cloud. Aristotle used an analogy with the situation in which the wind hits a narrow space in alleys and portals, or encounters wind from the opposite direction. In such situations, the wind is “thrust aside by the resistance either of the narrow entrance of the contrary wind and as a result forms a circular eddy of wind” (Aristotle 1952, Meteorologica, 370b23–25). The same mechanism occurs in clouds for the formation of tornadoes or waterspouts but combined with a downward motion. As stated by Wilson (2013), Meteorologica is not particularly clear on what produces the descent.
A tornado or a waterspout [whirlwind, in the original translation] thus arises when a windstorm [hurricane, in the original translation] that has been produced is unable to free itself from the cloud: it is caused by the resistance of the eddy, and occurs when the spiral sinks to the Earth and carries with it the cloud from which it is unable to free itself. Its blast overturns anything that lies in its path, and its circular motion whirls away and carries off by force anything it meets (Aristotle 1952, Meteorologica, 371a9–15).
This description is, to a certain extent, similar to the modern definition of a tornado, which the American Meteorological Society (2019a) defines as a rotating column of air, in contact with the surface, pendant from a cumuliform cloud (“spiral sinks to the Earth”) and often visible as a funnel cloud (“carries with it the cloud from which it is unable to free itself”) and/or circulating debris/dust at the ground (“its circular motion whirls away and carries off by force anything it meets”). Recent climatologies of tornadoes and waterspouts show their potential across much of Greece (e.g., Matsangouras et al. 2014; Avgoustoglou et al. 2018), suggesting that the ancient Greeks would have been familiar with both tornadoes and waterspouts.
Next, Aristotle introduced another type of weather phenomena that occurs “when the wind that is drawn down catches fire—which happens when it is finer in texture—[and] is called a firewind (—prēstēr); for its conflagration sets on fire and so colours the neighbouring air” (Aristotle 1952, Meteorologica, 371a16–19). The translation of prēstēr is not straightforward (Taub 2003). In the Greek–English Lexicon, the term is defined as a “hurricane or waterspout attended with lighting” (Liddell and Scott 1996). The term has been usually translated as whirlwind (e.g., Webster 1984; Wilson 2013), but Hall (1969) argued that presteros is the funnel-shaped body of a tornado or waterspout, and thus an accurate translation of prēstēr would be tornado or waterspout. For Aristotle, the wind is associated with presteros and also with keraunos (χεραυνóς, or thunderbolt), the difference between the two being the texture of the wind (i.e., the density of the material exhaled from the cloud). A thunderbolt is produced “if a large quantity of wind of fine texture is squeezed out of the cloud itself” (Aristotle 1952, Meteorologica, 371a20–21). Aristotle’s view that presteros and keraunos have a common origin is likely based on the observation that both tornadoes and waterspouts are associated with storms that produce lightning flashes (e.g., Peyraud 2013; Renko et al. 2013; Belo-Pereira et al. 2017; Stough et al. 2017). In the remainder of this article, presteros is translated as tornado or waterspout accompanied by lightning flashes. Thus, the theory proposed by Aristotle for the formation of tornadoes and waterspouts has three elements: 1) the wind is rotating inside the cloud and at some point is descending (generating windstorms), 2) the descending wind carries the cloud with it (generating tornadoes or waterspouts), and 3) the descending wind can catch fire (generating tornadoes or waterspouts accompanied by lightning flashes).
Aristotle’s successor in the Peripatetic School, Theophrastus of Ereus (c. 371–c. 287 BCE), is associated with several short works on meteorological subjects, like On Winds, Meteorology, and On Weather Signs (Daiber 1992). Theophrastus’s starting point for Meteorology was Aristotle’s Meteorologica. For Theophrastus, the exhalations used by Aristotle to explain the wind represent one of the possible causes of weather phenomena. According to Theophrastus, the wind moves in order to restore the balance disrupted by the sun (Taub 2003). Theophrastus indicated that a tornado or a waterspout accompanied by lightning is
Unlike Aristotle who was trying to provide single explanations for different weather phenomena, Theophrastus’s approach was to explain meteorological phenomena through multiple causes (Taub 2003). Thus, he described two mechanisms for the formation of tornadoes or waterspouts accompanied by lightning. The first mechanism produces a tornado or a waterspout when “a hollow cloud set itself up forcing the wind and stretches to the sea, the wind is pulled because of (the cloud)” (13.46; Daiber 1992). This mechanism suggests that an “empty” cloud is moving upward and thus is forcing the wind to rise with it and the cloud to stretch vertically. Through the second mechanism, which is similar with the one proposed by Aristotle, a tornado or a waterspout occurs when “the wind rushes off [the cloud] violently” (13.47; Daiber 1992).
Like Aristotle and Theophrastus, other natural philosophers speculated about weather phenomena, but some of their speculations were proposed as theoretical or practical explanations with the aim to eliminate the fear of deities (i.e., “attainment of an ethical aim”; Taub 2003). For example, Epicurus (341–270 BCE) tried to explain meteorological phenomena to eliminate the fear and anxiety associated with them (i.e., meteorological phenomena have a rational explanation and they are not caused by deities). Furthermore, Epicurus believed that multiple natural and logical causes should be considered for meteorological phenomena (Daiber 1992) and not causes based on divine intervention. In Letter to Pythocles dated from c. 305 BCE, Epicurus generally follows Aristotle’s explanations for the formation of the tornadoes or waterspouts accompanied by lightning.
Tornadoes or waterspouts accompanied by lightning [cyclones, in the original translation by Bailey (1926)] may be produced either by the driving down of a cloud into the regions below in the form of a pillar, because it is pushed by the wind gathered inside it and is driven on by the violence of the wind, while at the same time the wind outside impels it sideways; or by wind forming into circular motion, while mist is simultaneously thrust down from above; or when a great rush of wind takes place and cannot pass through sideways owing to the surrounding condensation of the atmosphere (Epicurus’s Letter to Pythocles 104.5–105.1, Bailey 1926).
The second explanation (“[…] or by wind forming into circular motion, while mist is simultaneously thrust down from above”) seems to suggest that tornadoes or waterspouts can form without a parent cloud. One possible interpretation of Epicurus’s text is that the wind has a circular motion over a certain height in the atmosphere. This circular motion is made visible due to the interaction with “mist […] from above” (or with cloud particles from the clouds above).
In summary, the earliest theories explaining the formation of tornadoes and waterspouts originate in Aristotle’s Meteorologica. His theories were later repeated by other philosophers (e.g., Theophrastus and Epicurus) who also proposed their own theories. The ancient Greek philosophy exerted a strong influence on the ancient Roman philosophy especially after 200 BCE when the Roman conquest of Greece began. One of the first Roman natural philosophers who wrote about meteorological topics was Titus Lucretius Carus (c. 99–c. 55 BCE), known simply as Lucretius, whose view on tornadoes and waterspouts is described in the next section.
4. Tornado and waterspout theories in ancient Rome
Unlike ancient Greece from where no records of tornadoes or waterspouts are known to exist, there are surviving reports of tornadoes (turbo in Latin)5 that occurred in the city of Rome and were described in Liber Prodigiorum (The Book of Prodigies) written around the fourth century CE by Giulio Ossequente (Ossequente 2005). The first known tornado report occurred in 152 BCE at Campo Marzio (today, the urban area between Capitoline Hill and Quirinal, two of the seven hills of Rome) close to the temple of Jupiter where the tornado demolished a column with its golden statue. The second tornado (60 BCE) occurred around noon and removed the roofs of houses, destroying a bridge (presumably a wooden bridge) and uprooting trees in the nearby countryside. The third tornado (44 BCE) detached the bronze plaques (i.e., plaques that displayed laws) from the Temple of Fede (Richardson 1992, p. 151), damaged the doors of the Temple of Ops (Richardson 1992, p. 277), uprooted trees, and destroyed many roofs.
The latter two of these tornadoes occurred during the lifetime of the Roman poet and philosopher Lucretius. Lucretius wrote his only known work in the first century BCE, the philosophical poem De Rerum Natura (On the Nature of Things), an account of the Greek Epicurean philosophy. Lucretius described in book VI of De Rerum Natura the core theories of Epicurean philosophy that were offered as an alternative to divine explanations of meteorological phenomena. Two mechanisms are presented by Lucretius for the formation of tornadoes and waterspouts:
[…] what the Greeks named presteres from above
Are sent down on the sea. Sometimes you know
A pillar, so to speak, is let right down
From sky to sea, round which the surges boil
Lashed by the blowing winds, and ships that are
Caught in that turmoil come in greatest risk.
And this takes place sometimes when the wind’s force
Cannot burst the cloud it aimed at, but can urge
It downward, like a pillar that is set
Tween sea and sky, coming by slow degrees,
Pushed and extended as’t were from above
Over the waves by strength of arm and hand:
And when the cloud is rent, the force of wind
Bursts forth upon the sea, and raises up
A wondrous surging in the waves around:
The eddy whirling round descends and brings
Yon cloud of pliant body down with it:
And having thrust it, heavy as it is,
Down to the level of the sea, the eddy then
Plunges itself entire into the waves,
And stirs the ocean with terrific noise,
And makes it boil. It chances too sometimes
That the eddying wind wraps up itself in clouds,
And gathering from the air the seeds of clouds,
As though let down from heaven, imitates
The prester. And when it has reached the earth
And burst, it vomits forth a whirling storm (turbinis)
Of vast dimensions, but as it is rare,
And mountains must obstruct its way on land,
More frequent it is seen in the wide expanse
Of ocean and beneath the spreading sky.
(Lucretius 1919, De Rerum Natura, book VI, 423–451).
The first mechanism explains the formation of waterspouts and is similar to the one proposed by Aristotle. Thus, the wind cannot break the cloud, and it is forced down in the shape of a pillar to the sea where it bursts and causes a furious boiling and surging. The second mechanism explains the formation of presteros (tornado or waterspout accompanied by lightning) and is similar to the second one proposed by Epicurus. In this explanation, the presteros forms outside the cloud by gathering the “seeds of clouds” (i.e., atoms of cloud) and wrapping them around. According to Lucretius, this type of presteros, formed through the gathering of “seed of clouds,” is sometimes observed over land (i.e., tornado), but often over the sea (i.e., waterspout).
Other major works of Latin literature include references to violent weather phenomena. In the Georgics, written around 29 BCE, the Roman poet Virgil (70–19 BCE) described what looks like the effects of a tornado or a waterspout.
[…] my own eyes have seen all the winds clash in battle, tearing up the heavy crops far and wide from its deepest roots and tossing it on high; then with its black whirlwind (turbine) the storm would sweep off the light stalk and flying stubble. Often, too, there appears in the sky a mighty column of waters, and clouds mustered from on high roll up a murky tempest of black showers: down fall the lofty heaven, and with its deluge of rain washes away the gladsome crops and the labours of oxen (Virgil 1916, Georgics, book I, 316–326).
At the beginning of the first century CE, two Roman authors wrote on meteorological subjects. The first author was Seneca (c. 4 BCE–65 CE), who around 65 CE wrote Naturales Quaestiones (Natural Questions), an encyclopedia of the physical sciences that included meteorological subjects. Taub (2003) argued that, unlike Epicurus and Lucretius whose aim in explaining the formation of meteorological phenomena was to free people from fear, Seneca had a genuine interest in meteorology as a physical science. For Seneca, a turbo forms when
[…] so long as there are no obstacles, [wind] gives its energy free rein; but when it is deflected by some prominence, or is forced into a sloping, narrow channel between converging land-masses, it wheels round on itself repeatedly and produces a vortex […]. This wind that revolves and circles on the same spot and whips itself up as it spins is a tornado or a waterspout6 (Seneca 2010, Naturales Quaestiones, book 5, 13.2–3).
Thus, for Seneca, the same mechanism that produces whirls in rivers (i.e., deflection of the course of the river by an obstacle) is responsible for forming whirlwinds in the atmosphere. If the whirlwind intensifies and lasts longer than ordinary, then “it catches fire and produces what the Greeks call a prester—that is, a fiery whirlwind (igneus turbo)” (Seneca 2010, Naturales Quaestiones, book 5, 13.3).
The second Roman author from the first century CE interested in meteorology was Pliny the Elder (23–79 CE), the author of Naturalis Historia (Natural History), an early encyclopedia dated c. 77–79 CE that used Greek and Roman sources. Pliny provided descriptions of meteorological phenomena and was also interested in ways of predicting those phenomena. Pliny described two types of winds, those that present some regularity and those that are sudden wind gusts. The sudden wind gusts are caused by Earth exhalations or by vapors associated with bodies of water (Taub 2003) and include the windstorm (ecnephias) and tornadoes and waterspouts (typhon) that form through the mechanisms proposed by Aristotle. In particular, a waterspout poses a threat to navigators “because [it] snatches things up and carries them back with it to the sky sucking them high above,” the only remedy being the “pouring [of] vinegar in advance of its approach, vinegar being a very cold substance” (Pliny the Elder 1967, Naturalis Historia, book 2, XLIX, 132). According to Pliny, tornadoes and waterspouts would never form when the wind gusts are from the north, and windstorms would never form during snowfall or when snow is on the ground (Pliny the Elder 1967, Naturalis Historia, book 2, XLIX, 134).
In summary, the Roman authors adopted the early theories proposed by Greek philosophers on tornadoes and waterspouts and further developed them. As we will show in the next section, the theories proposed by the Greek and Roman natural philosophers dominated meteorological thought for the next centuries.
5. Impact of the early theories on tornadoes and waterspouts through to the seventeenth century
Most of the theories on tornado and waterspout formation from ancient Greece and Rome can be traced back to Aristotle’s Meteorologica (Fig. 2). Almost all of the explanations from Meteorologica do not “even come close to the modern truth” (Wilson 2013, p. 2), and Aristotle “was wrong in profoundly interesting and significant ways”(Wilson 2013, p. 3). Regardless of the validity of Aristotle’s theories, he created a new scientific discipline (Lindberg 2008). Meteorologica not only influenced Greek and Roman natural philosophers, but it also influenced Muslim philosophers and scholars. Around 800 CE, the Antiochene scholar Ibn al-Biṭrīq translated and paraphrased Meteorologica into Arabic [see Schoonheim (2000) for a detailed discussion]. Over the next centuries, philosophers and scholars such as Ibn Sīnā (Avicenna; 980–1037) and Ibn Rušd (Averroes; 1126–98) wrote commentaries on Meteorologica and meteorological treaties inspired by Aristotle (Lettinck 1999). In Kitāb al-Šifā’ (The Book of Healing), Ibn Sīnā describes tornadoes and waterspouts (i.e., whirlwinds) as originating “[…] from the meeting of two strong (opposing) winds…Sometimes (the whirlwind) is firm (and) strong, (and) uproots the trees and wrenches ships from the sea. Sometimes (the whirlwind) encompasses a portion of cloud…and looks as if it were a dragon, flying in the atmosphere” (Sersen 1976).
Between the end of later Antiquity (c. 600 CE) and the early Middle Ages (c. 1100 CE), Aristotle’s writings were virtually unknown in western Europe. During this time, knowledge of the Greek language in western Europe declined as Latin grew (O’Hogan 2019). Also during this time, Isidore of Seville (c. 560–636) in Spain, the Venerable Bede (c. 673–735) in Great Britain, and Rabanus Maurus Magnentius (c. 780–856) in Germany were among the first medieval scholars to include meteorological topics in their works (Hellmann 1908). For example, in Etymologies (c. 600 CE), Isidore of Seville provided explanations for air, clouds, thunder and lightning, rainbows, and rain. He also mentions that tornadoes and waterspouts are produced by the “twisting of winds.”
Meteorologica was rediscovered in the twelfth century when Gerard of Cremona (c. 1114–87) translated into Latin the Arabic version by Ibn al-Biṭrīq. A new translation from Greek to Latin was completed in the second half of the thirteenth century by William van Moerbeke (c. 1215–c. 1286). Between the thirteenth century and throughout the Renaissance, Aristotle’s writings strongly influenced the physical sciences due to the “overwhelming explanatory power of [his] philosophical and scientific system” (Lindberg 2008). Furthermore, Aristotle’s view about nature and the divine were adopted by Christian philosophers and theologians during the Middle Ages (Grant and Martin 2017). There were also commentaries on Meteorologica, such as Roger Bacon’s (c. 1214–92) commentary in 1270, that were critical of Aristotle’s theories and that were perceived at that time as being close to blasphemy (Walker 2011).
Between the thirteenth and sixteenth centuries, very few works included discussions about the formation of tornadoes and waterspouts and their effects. One example of such a work that did discuss these whirlwinds is Historia de Gentibus Septentrionalibus (1555) (Description of the Nordic Peoples) by Olaus Magnus (1490–1557). In Historia de Gentibus Septentrionalibus, one of the first about Sweden and its inhabitants, Olaus Magnus dedicates a chapter to “the violence of the tornado (turbinis), and the storm” (Fig. 3). Olaus Magnus describes tornadoes as
A tornado (turbo) […] as Isidore says, is a gyration of winds, and is named from earth (terra); it occurs whenever the wind rises and drives particles of earth in a rotatory motion, the whirlwind being produced by several winds striving with one another, as Seneca testifies in his Naturales Quaestiones. It is generated and moves where there is land, and so tears up trees by the roots and strips bare whatever soil it falls upon; meanwhile, it snatches up woods and houses, travelling mostly below the clouds and certainly never higher. It is rounded in shape, and, rolling on like a spinning column, sweeps along faster than any cloud. Its movement is roving, discontinuous, and full of eddies; but it cannot last for long, for when a wandering and unsettled breath of air has formed a spiral, the total force eventually yields to one wind (Magnus 1996, Historia de Gentibus Septentrionalibus, chapter XI).
In the seventeenth and eighteenth centuries and even in the nineteenth century, authors generally repeated the theories proposed by Aristotle and Roman natural philosophers such as Lucretius7 (e.g., Tachard 1686; Hartsoeker 1696; Richard 1770; Page 1835). An interesting example is from Meteorologia Cosmica (Cosmic Meteorology) by Robert Fludd (1574–1637) (Fludd 1626) in which Aristotle’s theories about windstorms (ecnephias), tornadoes and waterspouts (turbo), and tornadoes and waterspouts accompanied by lightning (prester) are illustrated (Fig. 4). This figure shows that echoes of Aristotle’s views on tornadoes and waterspouts survived until the seventeenth century.
The period covering the end of the seventeenth century and the beginning of the eighteenth century marks the transition to a new period in the study of tornadoes and waterspouts in Europe. The theologian Francois Lamy (1636–1711) in France (Lamy 1689), the astronomer and mathematician Geminiano Montanari (1633–87) and the Jesuit polymath Roger Joseph Boscovich (1711–87) in Italy (Montanari 1694; Boscovich 1749) published book-length studies about waterspouts and tornadoes. These natural philosophers did not just simply repeat previous speculations about tornado formation, but in their attempts to understand this type of weather phenomena they also focused on providing detailed observations on tornadoes and waterspouts and their effects. Toward the end of the eighteenth century, new theories on tornado and waterspout formation began to emerge (Antonescu et al. 2016). One theory popularized by the French physicist Jean Charles Athanase Peltier (1785–1845), who also published the first pan-European tornado climatology, indicated that tornadoes are an electrical phenomenon (Peltier 1840). In the nineteenth century, the theories on tornadoes and waterspout formation proposed by Greek and Roman natural philosophers were cited only for their historical value. By the nineteenth century in the United States, these speculations about the formation of tornadoes and thunderstorms in general do not seem to have been considered by researchers in the field of severe storms. For example, in the Espy–Redfield debate about storm formation, no references are made, to our knowledge, to these earlier hypotheses (e.g., Kutzbach 1979). Sometimes the core ideas of modern theories were found in these speculations. Thus, the German tornado researcher Alfred Wegener (1880–1930) traced the mechanical and hydrodynamic hypotheses on tornado formation back to Lucretius and further to Aristotle (Antonescu et al. 2019). Even if the early ideas proposed by Greek and Roman natural philosophers about tornado formation were proven to be wrong by modern theories, summarizing and discussing these ideas contributes to a better understanding of the evolution of meteorological thought on severe storms.
We thank the University of Manchester Library for their help with the source material and for providing the image used in Fig. 3. Partial funding was provided to the University of Manchester by the Risk Prediction Initiative of the Bermuda Institute of Ocean Sciences through Grant RPI2.0-2016-SCHULTZ and the Natural Environment Research Council through Grant NE/N003918/1. Partial funding for Antonescu was provided by the AXA Research Fund to the University of Manchester for Assessing the Threat of Severe Convective Storms across Europe project. We also thank Konstantinos Fragkos from the Romanian National of Research and Development for Optoelectronics INOE2000 for assisting with the Greek translation. Authors Antonescu and Ene acknowledge funding of the Institutional Performance–Projects of Excellence Financing in RDI, Contract 19PFE/17.10.2018, of the National Core Programe Contract PN 18N/2019, and of the project Research Centre for environment and Earth Observation CEO-Terra, SMIS 470 code 108109, Contract 152/2016.
In this article, the term “theory” is used in its etymological sense as contemplation or speculation (from Ancient Greek ϑεoρíα—theōría) and not in the modern sense as “a supposition or a system of ideas intended to explain something, especially one based on general principles independent of the thing to be explained” (Oxford English Dictionary, s.v. “theory,” https://en.oxforddictionaries.com/definition/theory).
Before the nineteenth century when the term “science” began to be used with its modern meaning, “natural philosophy” was the philosophical study of nature and the physical universe.
According to Cannon and Kaye (1994), “typhoon” was transmitted by Cantonese via Arabic and, further, Persian, with its origin in Greek. The Chinese use of typhoon (i.e., tung fung—“a terrible storm of east winds” and ta fung—“a great wind”) is referring to a different type of wind than the Greek original.
In Latin, turbo is the nominative case and turbinis is the genitive case, which can be translated into English by the preposition “of” (e.g., “de violentia turbinis” can be translated as “the violence of the tornado”).
In the original Latin turbo; the English translation is whirlwind.
Considered lost, the poem “De Rerum Natura” was rediscovered in 1417 by Gian Francesco Poggio Bracciolini (1380–1459), an Italian humanist, historian, scholar in the Papal Court, and book hunter (Greenblatt 2012).