The bicentennial celebration of the inception of Frankenstein invites the present view of Victor Frankenstein and his fateful decision to destroy an unfinished female creature. The act itself was impulsive (caused by a “sensation of madness”), but it was preceded by agonized reasoning that would be familiar to any student of ecology or evolutionary biology. Here, we present a formal treatment of Frankenstein's reasoning and show that his rationale for denying a mate to his male creation has empirical justification. Our results suggest that the decision was prudent because it averted our own extinction by competitive exclusion. We ­conclude by suggesting that the central ­horror of Mary Shelley's novel lies in its ­prescient command of foundational concepts in ecology and evolution.

Some plot background

Victor Frankenstein created and then disavowed a nameless male creature described as 8 feet in height and proportionally large. For 3 years, the frightened creature wandered the European wilderness, becoming thoughtful, compassionate, and literate in three languages. In a pivotal scene, the creature encounters Frankenstein in Switzerland and pleads for a female companion “of the same species” to mitigate his loneliness. Crucially, and cleverly, the creature anticipates and preempts concerns of direct competition with humans; he promises geographic isolation and emphasizes resource partitioning:

If you consent, neither you nor any other human being shall ever see us again: I will go to the vast wilds of South America. My food is not that of man; I do not destroy the lamb and the kid to glut my appetite; acorns and ­berries afford me sufficient nourishment. My companion will be of the same nature as myself, and will be content with the same fare. We shall make our bed of dried leaves; the sun will shine on us as on man, and will ripen our food.

Frankenstein conceded to these assurances and commenced work on a female creature. However, he soon reflects on the potential for population growth and direct competition: “A race of devils would be propagated upon earth who might make the very existence of the species of man a condition precarious and full of terror.” The nature of this terror is clarified when he considers the probability of human extinction: “Future ages might curse me as their pest, whose selfishness had not hesitated to buy its own peace at the price, perhaps, of the existence of the whole human race.” Here, we indulge this anguished conjecture by asking whether and when a population of eight-foot creatures C could drive a population of humans H to extinction.

Modeling species interactions

To model species interactions, we elevated a tacit assumption to a formal parameter by assigning competitive advantages to creatures in a classic Lotka-Volterra competition framework. The effect of creatures on humans (which is always harmful by direct or indirect competitive interaction) is aHC, and the effect of humans on creatures is aCH, where aHC > aCH maintains a competitive advantage for creatures. Given growth rates for humans rH and creatures rC, as well as a carrying capacity for both k, the two-dimensional continuous time model is written  

\begin{equation*}\begin{array}{@{}*{1}{c}@{}} {\dot{H} = {r_H}H(1\,-\, (H\ + {a_{HC}}C)/k),}\\ {\dot{C}\ = {r_C}C(1\,-\, (C\ + \ {a_{CH}}H)/k),} \end{array}\end{equation*}
where human extinction is inevitable. However, the time to extinction could be as much as te= 108 years, which is tantamount to species coexistence at biological timescales.

Such an outcome is hardly surprising, given that the global population of humans in 1816 would have exceeded a founding population of two creatures by nine orders of magnitude. The population of Europe was then 178 ⋅ 106, whereas the global population was 1.01 ⋅ 109 (with an assumed carrying capacity of k = 1010). Given a human growth rate (= birth ratedeath rate) of rH= 0.0067, there is little reason for Frankenstein to envision imminent extinction. However, the creature is known to have recovered from a gunshot wound that “shattered flesh and bone,” suggesting that reanimated tissue is resistant to necrosis. If undead tissue dies at a slower rate than living human tissue does, then it follows that creatures should have a correspondingly lower death rate, such that the overall growth rate is rC= 1.5 ⋅ human growth rate.

These parameters shed new light on Frankenstein's decision to destroy his unfinished female creature. If we assume direct competition with humans in 1816 and if we allow the competitive advantage of creatures to vary from ε = 2⋅ to ε = 10⋅ the competitive effects of humans on creatures, then we can assess extinction time as a function of competition. In other words, aHC= εaCH, where the effect of creatures on humans is ε⋅ the effect of humans on creatures.

When competition is low, the time to human extinction is effectively infinite, meaning that populations of creatures and humans can coexist (figure 1a). However, as competition increases, our model shows that the time to human extinction drops precipitously to a minimum and then increases, an effect that becomes more exaggerated as the competitive advantages of creatures increase. Intriguingly, if the overall level of competition is high, the creatures are doomed to extinction despite their competitive advantage, such that the time to human extinction is also infinite. This result occurs because the population of creatures begins at n = 2 individuals, a population size that is too small for establishing a competitive foothold. In the worst-case scenario for humans (aHC = 3.5, ∑ = 10), our global model indicates human extinction in te= 4188 years. We invite readers to explore simulations and model results with an interactive Mathematica notebook, available in the supplemental materials.

Figure 1.

(a) Extinction time given a competitive environment. (b) Population trajectories for humans (descending) and creatures (ascending) given initial growth in the Amazon catchment (dashed) and Europe (solid). (c) Time to extinction versus the human population size during initial growth of the creature population.

Figure 1.

(a) Extinction time given a competitive environment. (b) Population trajectories for humans (descending) and creatures (ascending) given initial growth in the Amazon catchment (dashed) and Europe (solid). (c) Time to extinction versus the human population size during initial growth of the creature population.

Dispersal to South America

Given the large demographic disparity between creatures and humans, it is worth considering whether different environmental parameters might alter competitive outcomes. Recall that the creature promised to inhabit “the vast wilds of South America” in an apparent gesture of conciliation. We therefore explored the effects of dispersal to South America by comparing interactions between creatures and humans in the Amazon catchment (dashed curves, figures 1a and 1b) and Europe (solid curves, figures 1a and 1b). We assume that when a population of creatures reaches 90 percent of the carrying capacity of either environment, it will begin direct competition with the global human population.

Mount Tambora and the year without summer.

Recent tributes have renewed interest in the eruption of Mount Tambora in April 1815 and its pernicious aftereffects, particularly the anomalous weather of 1816 (the “year without summer”). The paroxysmal eruption was cataclysmic, ejecting nearly 175 cubic kilometers of volcanic debris, including 50 million tons of sulfur dioxide, which rose into the stratosphere, enveloped the Earth, and oxidized to form small, light-reflecting sulfate particles. The reflection of sunlight reduced the energy absorption of Earth and caused cooler temperatures. In the Northeast United States, summer snow and an unrelenting series of August frosts destroyed crops, caused famine, and gave rise to the colloquialism “eighteen-hundred-and-froze-to-death.”

Darker skies also contributed to the gloom, conditions that Lord Byron described beautifully as “despairing light” of “mad disquietude.” These words appear in one of his most celebrated poems, fittingly titled “Darkness.” The bleak mood of 1816 is familiar to students of Victorian literature for its influence on many writers, including Byron and Percy Bysshe Shelley, who were then visiting Lake Geneva, Switzerland, with a literary coterie that included Mary Shelley, Mary's stepsister Claire Clairmont, and the physician John Polidori. The inclement conditions drove the party indoors; it was, in Mary's words, “a wet, ungenial summer [of] incessant rain.” Fireside gatherings in the Villa Diodati (pictured) led to the reading aloud of ghost stories and Byron's challenge to each person to write their own ghostly tale. For Mary, the proposition provoked a “waking dream” (probably 16 June 1816) that eventually gave rise to Frankenstein, published in 1818, whereas John Polidori produced The Vampyre, published in 1819. The characters and gothic tone of these works have had a large influence on popular culture, a volcanic byproduct that continues to resonate two centuries on.


Photograph by Russell L. Doescher, reproduced with permission.

Photograph by Russell L. Doescher, reproduced with permission.

A founding population of two creatures in South America (ascending green dashed line, figure 1b) would quickly surpass the local human population (blue dashed line, figure 1b; the vertical jump denotes the transition from regional to global competition) and drive humans to extinction faster than competition in Europe (creatures: ascending green solid line; humans: blue solid line; figure 1b). The lower te for this scenario holds across all levels of competition and for all values of ε (dashed curves, figure 1a). The wilds of South America would therefore accelerate the population growth of creatures, at least compared with the slower dynamics that would have occurred in Europe, with its larger human population. In general, we find that low-density—and, by extension, low-competition—initial environments can catalyze the establishment of an invading population, thereby hastening the extinction of a resident population (figure 1c).

The present findings are drawn from a work of science fiction, but their importance is threefold. First, our results reinforce and expand the gothic tone of Frankenstein and its underlying exploration of moral and scientific responsibility. Second, our results cast new light on the creature and his motives for inhabiting the wilds of South America, a lower-competition environment. Third, our results bolster the speculative concerns of Frankenstein with empirical support: Humans would indeed face species interactions “full of terror.” The nature of this terror is termed competitive exclusion, a concept that escaped definition until the 1930s. We conclude by suggesting that the central horror and genius of Mary Shelley's novel lie in its early mastery of foundational concepts in ecology and evolution.

Further reading.

Aldiss BW. 1986. Trillion Year Spree: The History of Science Fiction. Victor Gollancz.

D'Arcy Wood G. 2014. Tambora: The Eruption that Changed the World. Princeton University Press.

Hardin G. 1960. The competitive exclusion principle. Science 131: 1292–1297. doi:10.1126/science.131.3409.1292

Holmes R. 2016. The science that fed Frankenstein. Nature 535: 490–492. doi:10.1038/535490a

Luterbacher J, Pfister C. 2015. The year without a summer. Nature Geoscience 8: 246–248. doi:10.1038/ngeo2404

Olson DW, Olson MS, Doescher RL, Pope AG, Schnarr KD. 2011. The moon and the origin of Frankenstein. Sky and Telescope 122: 69–74.

Oppenheimer C. 2003. Climatic, environmental, and human consequences of the largest known historic eruption: Tambora volcano (Indonesia) 1815. Progress in Physical Geography 27: 230–259. doi:10.1191/0309133303pp379ra

Post JD. 1977. The Last Great Subsistence Crisis in the Western World. Johns Hopkins University Press.

Shelley MW. 1818. Frankenstein, or The Modern Prometheus. Lackington, Hughes, Harding, Mavor, and Jones.

Stableford B. 1995. Frankenstein and the origins of science fiction. Pages 46–57 in Seed D, ed. Anticipations: Essays on Early Science Fiction and its Precursors. Syracuse University Press.

Supplemental material

Supplementary data are available at BIOSCI online.


This Viewpoint stems from an assignment in an undergraduate course, Victorian Literature and Science, taught by Professor Mary Poovey in 1996. We are grateful for her feedback then and for the recent comments of Christopher Kempes, Taal Levi, Eric Libby, Marc Mangel, Patricia McKee, and Mark and Sarah McPeek.