It Wasn't The Sun!

Reply to Scott Alexander

0. Introduction

Scott Alexander replied to my article on Our Lady of Fatima. Although he remains skeptical, his analysis was actually quite favorable to authenticity. He acknowledged that he is strongly motivated to reject Catholicism: “I hope the visions of Fatima were untrue, and therefore I must also hope the Miracle of the Sun was a fake. But I’ll also admit this: at times when doing this research, I was genuinely scared and confused.” After he finished laying out his case, Scott concluded that “I don’t think we have devastated the miracle believers. We have, at best, mildly irritated them. If we are lucky, we have posited a very tenuous, skeletal draft of a materialist explanation of Fatima that does not immediately collapse upon the slightest exposure to the data.”

In this reply, I argue that Scott’s tenuous, skeletal draft of a materialist explanation of Fatima immediately collapses upon the slightest exposure to the data. In the first section, I respond to Scott’s discussion of cloud dimming. In the second section, I respond to Scott’s discussion of the distant witnesses. In the third section, I respond to Scott’s discussion of the photogrammetric evidence. In the fourth section, I respond to Scott’s discussion of rapid drying. In the fifth section, I respond to several objections that I didn’t have the opportunity to address in the preceding sections.

1. Cloud Dimming

The consensus of witnesses—including of those that gave ‘negative testimony’—was that the ‘Sun’ lost its dazzling brightness, was perfectly comfortable to stare at, and presented as a sharply bounded disc throughout the Miracle of the Sun.

Despite the fact that “every previous commentator acts as if this is some cosmic mystery,” Scott believes that these phenomena can be explained by mundane cloud dimming. However, that hypothesis contradicts all of the witnesses and his own explanation for the rest of the phenomenology that was reported by the crowd.

1.1 - Testimony

Scott appeals to Catholic physicist Fr. Stanley Jaki’s observation that “the great majority of eyewitness accounts, and certainly the most important ones, contain emphatic references to the continued presence of clouds.” Scott neglects the fact that those ‘emphatic references’ both explicitly and implicitly contradict his proposal.

• Sampling from Scott’s collection of testimonies from 60 eyewitnesses, I found 15 statements that unambiguously describe the behavior of clouds during the event. All of them confirm that, although clouds were present and sometimes passed in front of the ‘Sun,’ cloud coverage was partial, nonuniform, and intermittent.¹

  • Jose Garrett gave the most detailed, technically articulate, and judicious account that is preserved in extant records.

    • Garrett explicitly considered and rejected cloud dimming: “It couldn’t be confused with the sun seen through fog (which, by the way, didn’t exist at that time) because it wasn’t opaque, diffuse, or veiled.”

    • He also reported that cloud coverage was intermittent: “the sun sometimes stood out in patches of clear sky. The clouds that ran lightly from west to east did not obscure the light (which did not hurt) of the sun… but sometimes these flakes, which were white, seemed to take on, sliding before the sun, a diaphanous pink or blue hue.”

  • Mario Godinho reported “I saw in a clear area of sky (where one should not be able to stare at the sun) the very sun… I could then see the sun more easily than I can see the moon on a full moon night.”

  • Avelino de Almedia wrote “one could see the immense crowd turning toward the sun, which appears free from all clouds and at its zenith. The sun resembles a dull, silver plate and it is possible to look at it without the least effort. It does not burn, nor blind. It seems as if an eclipse is taking place.”

  • Joaquim Vicente reported “It was raining. Suddenly the clouds opened and the sun seemed as though through a window.”

  • Jose Curado reported “The sun, now cloudless, did not project its rays onto the earth… Suddenly, it lost all light or flames and a pale silver plate appeared… It could remain fixed indefinitely without disturbing the eyes.”

  • Antonio Marques reported “The sky was covered with clouds, but suddenly the clouds tore away and the sun appeared as if trembling.”

  • Jacinto Lopes reported “The hour approaches, and behold, as if by magic, the rain stops, the sun breaks through the dense, black clouds and reveals itself with luminous rays… and soon loses its brightness and colors—able to be seen by the naked eye without hurting the eyes…”

  • Dominic Reis reported “Yes, I could look at the sun without pain in the eyes… There were colors all around the sun… and it was like the real sun… but the clouds didn’t go with the sun… they were blue and different colors…”

  • Raio de Luz reported “Suddenly, a luminous disc the size of a large host, but which everyone could gaze upon as one gazes upon the moon, appeared, as if it had detached itself from the sun (or rather, placed itself before it)… a beautiful cloud, now fiery-colored, now purple, now pink, now golden, appeared and disappeared successively…”

  • Jose Silva reported “We could look at the sun without difficulty, the sky became clear…”

  • Luis Vasconcelos reported “I was absolutely convinced that I would see nothing… at first I only saw clouds drifting by, leaving the sun uncovered. Suddenly, I see an intensely pink rim, surrounding the sun, which resembled a disc of dull silver, as someone once said, while giving me the impression that it was moving from its original position. Diaphanous, vaporous clouds, somewhat purple, somewhat orange, permeated the air.”

  • Joseph Frazao reported “The sky was covered with clouds but suddenly the clouds opened.”

  • Higino Faria reported “At one o’clock the clouds gathered into a very thick and dark form, giving the appearance of an eclipse… Then the dark cloud broke into parts, and through the break we saw the sun shining…”

  • Maria Queiros reported “Minutes later, after the arrival of the little shepherds, the rain stopped, and the sky, which had been dark until then, turned precisely at the time… at one o’clock, the sky, where the cloud has strayed, cleared; and what was our surprise when a silvery globe appeared, appearing to be crossed here and there by the clouds…”

  • Maria Teresa reported “We could not see the sun. Then suddenly, at noon, the clouds drew away and the sun appeared as if it were trembling… we could look at it as if it were the moon.”

1.2 - Inconsistency

The vast majority of witnesses reported anomalous color and motion perception. Scott thinks his “most promising lead” is that similar phenomena have been reported by sungazers. However, the sungazing effects are physiologically contingent on retinal overstimulation by an extremely bright, high contrast source. That is why entoptic phenomena are sometimes reported by sungazers but never by ‘moongazers.’

The function of discomfort and accompanying reflexes is to prevent the visual system from entering regimes in which perceptual distortions arise. That puts Scott in a double bind. If he invokes attenuation to explain prolonged, comfortable fixation, then he excludes the perceptual mechanisms that he needs to accommodate the rest of the data. Conversely, if he maintains that the source was intense enough to trigger entoptic phenomena, then he can no longer accommodate comfortable fixation.

Scott tries to thread the needle by proposing “some previously unknown optical illusion that potentially causes the sun to appear to change colors and spin” that is “somehow modulated by cloud cover.” He suggests that “when the sun is half-hidden by translucent clouds, the illusion happens much faster.” That is wildly implausible:

• First, it is not clear what Scott means by ‘half-hidden.’ If he means uniform veiling, then dimming and dulling would have inhibited rather than promoted sun-gazing effects. If he means nonuniform veiling, then luminance gradients would have made discomfort glare even worse than it would have been in a clear sky. If he means inconstant veiling, then temporal fluctuations in luminance would have caused flicker glare which is significantly worse than static glare.

• Second, color illusions occur because the visual system interprets ambiguous stimuli in ways that minimize surprise, suppress contrasts, and maintain color constancy. Each of those tendencies opposes illusions of alternating, saturated hues that have no spectral basis. Distortions from sungazing are better understood as physiological malfunctions than as perceptual illusions. Those aren’t modulated by clouds and expectancy in the way Scott needs them to be.

• Third, several witnesses saw alternating colors without looking at the ‘Sun.’ Carlos Silva reported that he “did not look at the sun to avoid the optical illusions… and nevertheless saw different colors in the faces of the huge audience, and especially in the white felt hat of a lady...” Manuel Gonçalves Junior reported that “with his eyes always fixed on the holm oak… without deliberately looking at the sun, noticed that both the people and the trees and everything that his vision reached in the direction of the holm oak, took on different colors.”²

• Fourth, apparent motion is viewpoint-dependent. No single effect would produce convergent sequences of illusory rotational, horizontal, and vertical motion for observers viewing a source from a wide range of angles, distances, and elevations. When sungazers report apparent motion, it is always idiosyncratic. To accommodate the degree of convergence between witnesses, Scott would need to posit a mechanism that was closer to mass psychosis than to optical illusion.

2. Distant Witnesses

2.1 - Alburitel

Scott writes “The harder I looked, the more Alburitel testimonies I found, until I ended up with three. All three seem to be talking about the same event - a crowd who gathered outside the school to watch the phenomenon.” Actually, two groups witnessed the Miracle of the Sun from separate locations in the vicinity of Alburitel.

• (Hilltop) A group of about 20 villagers ventured off to a nearby hill from which they could see, far off in the distance, the ridges of Fatima. This group included Joaquim Lourenço who provided a detailed account of his experience:³

A group of villagers gathered and, despite the bad weather, went up to a hill from which one commands a vast horizon, seeing, very far off, the ridges of Fátima. That group waited there until midday. Among the bystanders was the present Rector of the Sanctuary of Fátima, Rev. Dr. Joaquim Lourenço, elder brother of the aforementioned missionary Rev. Father Inácio Lourenço, who accompanied their mother. His Reverence remembers perfectly what took place — something both pathetic and dramatic — on that hill at that hour.

Suddenly the sun broke through the clouds and appeared in a dizzying rotation, then seemed, in an impetuous descent, to tear itself loose from the sky to fall upon the earth and kill everyone. Those people — about twenty persons — began shouting loudly, begging God and Our Lady for help, thinking they were about to die. And when the sun made a leap toward them, the panic could not have been greater. They threw themselves on the ground and awaited the crash…

The Rev. Dr. Joaquim Lourenço told me last year that, having thrown himself to the ground and finding that death did not come, he raised his head and saw that the sun had gone back up and was again revolving on high, emitting, without hurting the eyes, a multicolored light — something never before seen. That solar leap happened three times, and each time the people threw themselves to the ground, crying out, thinking it was the end of their lives. Then everything returned to normal in the sky. The people went home stupefied.

• (Village) Meanwhile, the other group stayed near the village center of Alburitel. That group included Inácio Lourenço, then a schoolboy, who later wrote a letter that described what happened. His account was corroborated by his teacher.⁴

Inácio Lourenço was a nine-year-old boy at the time, living in the village of Alburitel, ten miles away from Fatima. He is now a priest and he remembers this day vividly. He was in school. “About noon,” he said, “we were startled by the cries and exclamations of the people going by the school. The teacher was the first to run outside onto the street with all the children following her. The people cried and wept on the street; they were all pointing towards the sun. It was ‘The Miracle’ promised by Our Lady. I feel unable to describe it as I saw it and experienced it at the time. I was gazing at the sun. It looked so pale to me; it did not blind. It was like a ball of snow rotating upon itself. All of a sudden it seemed to be falling, zigzag, threatening the earth. Seized with fear, I hid myself among the people. Everyone was crying, waiting for the end of the world.

“Nearby, there was a godless man who had spent the morning making fun of the simpletons who had gone to Fatima just to see a girl. I looked at him and he was numbed, his eyes riveted on the sun. I saw him tremble from head to foot. Then he raised his hands towards Heaven, as he was kneeling there in the mud, and cried out, ‘Our Lady, Our Lady.’ Everyone was crying and weeping, asking God to forgive them their sins. After this was over, we ran to the chapels, some to one, others to the other one in our village. They were soon filled.”

“During the minutes that the miracle lasted, everything around us reflected all the colors of the rainbow. We looked at each other and one seemed blue, another yellow, another red, and so on. This increased the terror of the people. After ten minutes, the sun resumed its place, pale and without splendor. When everyone realized the danger was over, there was an outburst of joy. Everyone broke out in a hymn of praise to Our Lady.”

Scott notes that the villagers in Alburitel “anticipated that the miracle would involve the stars,” so they had been “darkening bits of glass by exposing them to candle-smoke so that they might watch the sun with no harm to their eyes.” He concludes that witnesses in Alburitel were ‘contaminated’ by expectation. However, that does not even begin to explain the most significant implications of their accounts:

• First, the Cova and Alburitel look through independent atmospheric corridors when viewing the Sun. If the gazed source was the Sun, it would have to be an extraordinary coincidence that an extremely rapid clearing in Fatima’s sightline occurred at approximately the same time as an extremely rapid clearing in Alburitel’s sightline. By contrast, if the gazed source was a localized emitter due south of the Cova, then the same break could have opened both sightlines.

• Second, Leonor de Avelar e Silva Constancio reported that there was “light, persistent rain” around Torres Novas during and after the Miracle of the Sun. Torres Novas lies due south of Alburitel, directly along the solar azimuth. In other words, it occupies the same atmospheric corridor that Alburitel would have had to look through to see the true Sun. It is very unlikely that Alburitel observers had a line of sight to the true Sun if there were storm clouds over Torres Novas.

• Third, the ‘expectation’ of the villagers was merely that the miracle would ‘involve the stars.’ That is not specific enough to account for the degree of convergence between the testimonies from the Cova and from in and around Alburitel. The crowds were too far apart to coordinate their experiences during the event, yet witnesses from all three reported almost exactly the same course of events: the spontaneous emergence of a luminous disc that could be stared at without discomfort, its rotational and translational motion, its dramatic threefold descent towards the earth, and multicolored light that reflected off the environment.

2.2 - Minde

Scott writes “From the child, Albano Barros, we have only two sentences, not enough to know whether he was contaminated or not. But he was nine years old, and his account was collected thirty years later.” Actually, Barros was twelve years old in 1917.

Albano Barros reported “I was watching sheep, as was my daily task, and suddenly there, in the direction of Fatima, I saw the sun fall from the sky. I thought it was the end of the world… I was so distracted that I remember nothing but the falling sun.” Crucially, he also recalled that “almost everyone in the area thought the world was ending.”⁵ Even though Barros’ testimony is concise and late, it has evidential value:

• First, Barros noted that he was looking ‘in the direction of Fatima.’ Since he was in Minde, that implies he was facing the opposite direction of the Sun at the time of the event. That detail is unlikely to be explained by memory conformity, since the prevailing interpretation among uneducated witnesses after the event was that the miracle was astronomical. It is possible that it is explained by memory drift, but it contributes to an undesigned coincidence among distant witnesses.

• Second, Barros implied that he was not staring fixedly at the ‘Sun’ when he saw it ‘fall from the sky.’ He claimed the event intruded on his normal activity—tending sheep—and that his distress prevented him from attending to any secondary phenomena. It is suggestive that the only detail he remembered is the one that would have made it possible for the event to abruptly enter into his visual field.

• Third, Barros characterized the attitudes of ‘almost everyone in the area.’ That implies that this was a communal experience that would have been repeatedly discussed by the villagers in Minde for years to come. That is compelling evidence there were many witnesses in Minde, since it is unlikely that Barros confabulated that most of his family and neighbors claimed to have witnessed a cataclysmic, world-historical event in the immediate vicinity of their hometown.

2.3 - Leiria

Scott writes “Moving on to the housewife, Guilhermina Lopes da Silva, her story seems real enough, but she tells us that ‘I could not go [to Fatima] because my husband was an unbeliever.’ She knew a miracle was predicted, wanted to see it, but had to stay home… I don’t think that she can be described as uncontaminated.”

Guilhermina Lopes da Silva reported “I was looking toward the mountain at noon when suddenly I saw a great red flash in the sky… I called two men who were working for us. They, of course, saw it, too.”⁶ Once again, ‘contamination’ by expectation can’t explain the most important features of this testimony:

• First, Silva was looking towards the mountain. The Serra de Aire e Candeeiros lies to the southeast of Leiria, interposed between the city and the Cova. Meanwhile, the solar azimuth would have been due south at the time. The phenomenon she described—’a great red flash in the sky’—is not consistent with staring at the Sun. Scott has to assume that Silva, as well as the two workmen, experienced perceptual anomalies that coincided exactly in time with independent anomalies reported by the crowds that were proximate to the Cova.

• Second, a ‘red flash’ is exactly what you would expect from long-distance scattering of multicolored light. At a grazing angle, the path of light from the Cova to Leiria would have traversed dozens of kilometers of dense air. Shorter wavelengths—blue, green, and yellow—would have been scattered out of the direct beam, leaving only long-wavelength red light to reach the observer. The same principle explains why the Sun turns deep red at sunrise and sunset.

Silva’s testimony is corroborated by a newspaper article that was published on November 5, 1917. Fr. Jose Ferreira de Lacerda wrote “How was it possible for there to be suggestion among people of such diverse ranks, and even more so among people who, not having been to Fátima, witnessed the event in Leiria?” That suggests that other witnesses in Leiria had a better line of sight to the Cova than Silva.

2.4 - Sao Pedro de Muel

Scott writes “Moving on to the poet - Alfonso Lopes Vieria by name - this testimony is on shakier ground. We hear it secondhand, from the writer of a book on Fatima who claims to have interviewed the poet almost twenty years later - and then from a confirmation by his widow thirty years later, who told another writer that yes, he definitely said it.” Actually, Vieria’s widow did not merely confirm that ‘he definitely said it’ — she confirmed that she personally witnessed the event with him.

Alfonso Lopes Vieria reported that “While working on the veranda at noon, on October 13th, 1917, he was surprised by the astounding solar phenomenon then taking place at the Cova da Iria. He enthusiastically called to his wife and mother-in-law to come and see. With deep conviction, he told his friends about what he had seen, and his widow Senhora Helena de Aboim Lopes confirms his words.”⁷

Helena de Aboim Lopes reported that “having hurriedly followed the Poet with the other people present to the balcony, they observed from there the marvelous spectacle which they later learned had occurred in the Cova da Iria.”⁸

Scott raises three objections to this testimony. I believe they are far too weak to justify the inference that both Alfonso and Helena were lying or confabulating:

• Scott writes “The poet says in his testimony that he had ‘no recollection’ that a miracle was planned for Fatima that day. But it was all anyone was talking about, for weeks, and the roads the previous day would have been choked with pilgrims heading for the town. How did he not know about this?”

  • ‘It was all anyone was talking about for weeks’ and ‘the roads the previous day would have been choked with pilgrims’ are gross overstatements. Diario de Noticias—Portugal’s largest newspaper—did not report on expectation of a miracle October 13th until the day after. O Seculo only ran two short and derisive articles before October 13th. Leiria papers ran stories about the apparitions, but their circulation was limited. Most of the pilgrims came from areas that were along the Lis–Zezere valley routes. There was not a massive exodus of pilgrims from seaside villages on the Atlantic coast.

• Scott writes “After the miracle, the local bishop put out a request for witnesses, especially those… who had seen it from a distance. The poet’s testimony would have been priceless. But although the poet wrote the bishop about other matters during the interim, he does not mention having seen the miracle, and the bishop seems unaware of him in his summary of the evidence.”

  • The Bishop of Leiria did not refer to specific witnesses in his ‘summary of the evidence.’ However, he wrote “This phenomenon, which no astronomical observatory recorded and was therefore unnatural, was witnessed by people of all social classes and backgrounds, believers and nonbelievers, journalists from major Portugese newspapers, and even individuals miles away…”⁹ It is entirely possible that Vieria spoke to the bishop in person, some of his letters to the bishop were lost, or that he was negligent or bashful. Fr. Jaki’s book is littered with examples of surprising silences by known witnesses.

• Scott writes “The poet also corresponded with Manuel Formigao, a man who spent his life gathering evidence for the Miracle of the Sun... But he never bothered to say “Hey, you know the thing you’ve devoted your life to? I have priceless evidence that it’s completely real. He just mentioned other things…”

  • Fr. Jaki references a single letter between Vieria and Formigao where “the poet’s concerns were much more about style than about facts.” Elsewhere, Fr. Jaki writes “Canon Formigao did not even do what he most certainly could have done with the help of Prof. Garrett, namely, contact other educated people for eyewitness accounts about the event witnessed in the Cova…”¹⁰ It is foolish to discount multiply attested testimony based on arguments from silence that are themselves based on inferences from fragmentary records.

2.5 - Beaches

While researching this article, I found examples of distant witnesses that have been overlooked by the secondary literature on the Miracle of the Sun.

• (Santo Amaro de Oeiras) Berta Leite, daughter the painter Jose Leite, wrote:¹¹

The miracle of Fatima that my father saw was on 13 October. He liked to take the train to Santo Amaro de Oeiras and spend some hours painting... He went to paint the rocky shore at the beach, and suddenly had to stop—the sun was turning... He came home enchanted... and said: “It was beautiful! It was beautiful!”

• (Praia da Granja) Maria Garrett was at the Cova for the Miracle of the Sun. The next day, on October 14th, she wrote the following in a letter to Dr. Formigao:¹²

here at Granja… there were people who saw the movement of rotation of the sun and the rose-colored cloud pass before the sun.

Santo Amaro de Oeiras is 120km south of the Cova. Praia da Granja is 160km north of the Cova. Nobody that far away would have been expecting to see a miracle. Scott concedes that his theory “can’t explain true uncontaminated distant witnesses.”

2.6 - Torres Novas

Scott writes “we have at least one explicit negative distant witness. This is Leonor de Avelar e Silva Constancio, who we met before - she missed the miracle when her car got stuck in a ditch a few miles outside Fatima…. If [the event] was visible within a twenty miles radius, she shouldn’t have missed the miracle at all!”

• First, Constancio was in terrible viewing conditions. Her car had slid into a ditch and it was raining. The near-field landscape around Torres Novas is cluttered with occluders—embankments, trees, and rising ground—that truncate sightlines. From that vantage point, under a thick overcast and amid downpour, it is highly probable that Leonor didn’t have a view of the horizon to the northwest.

• Second, inattentional blindness is real and powerful. Psychological studies have demonstrated that people routinely fail to perceive dramatic stimuli when their attention is directed elsewhere. Leonor didn’t anticipate a sky miracle, she didn’t know the exact time the event would occur, and she would have been distracted by the fact that her car had broken down. Even if the light source were visible from her vantage point, it would have been entirely possible for her to miss it.

Scott writes “if we believe Dalleur’s location, we can use trigonometry to estimate the light source’s elevation at >1 mile. This could not have been blocked by the small hills near Torres Novas, and so the explicit negative evidence from Constancio - not to mention the implicit negative evidence from the other 40,000 residents of Torres Novas - becomes damning.”

• First, what matters is not whether the source cleared the ridge line, but whether it would have been visible through storm clouds and near-field occluders. A localized emitter near the Cova that was between 1.5-3 kilometers above sea level would have appeared 4.5° to 8° above the horizon for observers in Torres Novas—a very low angle, easily obstructed by terrain, vegetation, and clouds.

• Second, there is not ‘implicit negative evidence’ from the other residents of Torres Novas. Adelaide Grego wrote “Here in Torres Novas at the same time something also manifested in the sun, but from the description they give it cannot be compared.”¹³ That confirms anomalous phenomena were observed by witnesses in Torres Novas, albeit less distinctly than in Alburitel or Minde.

2.7 - Too few?

Scott writes “there were about 300,000 people living within a 20 mile radius of Fatima in 1917. If 50,000 of those had gone to Fatima itself, and another 100,000 were in the southeast area blocked by mountains, then 150,000 people outside Fatima still should have seen the miracle. Of those 150,000, we have four to six testimonies - compared to 100+ testimonies from the mere 70,000 at Fatima itself. Is this surprising?”

Scott concedes “Maybe not: It was a rainy day; many people stayed inside. And the event might have been very dramatic at Fatima, but only slightly visible… elsewhere. Maybe you had to be outside in the rain, staring directly at the right part of the horizon, and not that many people were in that category.” However, he goes on to raise the objection that “Dalleur must believe that the event seemed cataclysmic up to at least a 10 mile radius. So where are all the other distant witnesses?”

• First, based on the 1911 Portugese census, the population of the 15km radius around Dalleur’s localized emitter would have been approximately 40,000.¹⁴ A great majority of the population in that area would have been in towns with terrible sightlines. A substantial fraction of the remainder would have been attendance at the Cova. Given the weather, it is entirely possible that no more than a few hundred people would have seen the emitter from within that radius.

• Second, there is a sampling bias. Fr. Jaki complains that canonical investigations failed to systematically elicit and record testimony. Scott notes that “our best source for witness testimonies is the Documentacao Critica de Fatima.” The excerpts that have been digitized include the Bishop of Leiria’s reference to “individuals miles away,” but do not include testimony from any of them. The accounts that have been preserved were either written in letters or elicited by John Haffert decades later, when the task would have been much harder.

The geographic distribution of the distant witnesses that we do have testimony from is exactly what you would expect from a localized emitter. In Minde, which would have been as close to the emitter as the Cova, “almost everyone in the area thought the world was ending.” In Alburitel, which had an elevated vantage and an unobstructed horizon, witnesses from two separate groups reported exactly the same phenomena as the witnesses in the Cova. In Torres Novas, located at roughly the same distance as Alburitel but with much worse visibility, witnesses reported anomalous phenomena, but not to the same extent as Alburitel. In Leiria, much farther away, witnesses perceived a fingerprint of long-distance scattering of multi-colored light. Finally, all of the most remote witnesses were in coastal areas that would have had good visibility. Rotational motion and a reddened diffuse glow were the two features of the event that would have been the most likely to survive over long optical paths.

3. Photogrammetry

Dalleur contends that it is possible to exclude a solar identity for the primary light source (LSa) of a certified photograph (D115) of the crowd at the Cova during the Miracle of the Sun by inferring its angular elevation from shadows. Scott defers to Mark Grant’s rebuttal. At the outset, it is worth noting that, whereas the Shrine gave Dalleur access to a 6k scan, Mark told me that they only gave him access to a 4k scan.

Mark acknowledges there is an unambiguous shadow beneath a sharp rock in the bottom-left corner. The lip of the rock and the ground immediately below are clearly distinct surfaces. The boundary between light and shade is linear and hard-edged, showing minimal diffusion. The receiver’s surface is nearly planar across the visible run, so parallax and surface curvature introduce almost no distortion. In other words, the shadow behaves as an ideal 3D projection of the illumination vector.

However, Mark questions the reliability of the other shadows that Dalleur relies on:

• (Hat brim) Dalleur derives one of his principal beam lines from a sharply defined dark band running along the underside of a boy’s hat brim. He interprets this as a tangent self-shadow cast by the crown onto the brim—an established geometric feature of fedora-style hats, where the illumination ray grazes the lower rim of the crown and meets the brim at a tangent. Mark worries that the shadow may spill onto the boy’s leg, which is curved and non-planar, in which case the tangent construction would lose heuristic validity. Against this, the shadow clearly aligns with the expected crown–brim junction, remains straight across its length, and falls short of the outer edge of the brim. Even if a continuation did extend onto the leg, its contribution would be too short and oblique to alter the direction by more than a few degrees—within the tolerance of Dalleur’s analysis.

  • There are three supplementary cues in the vicinity of the brim shadow. A sharp cast shadow from the brim’s lower rim falls on a light, nearly planar background, giving a direct geometric measure of the projected beam on a surface independent of the hat itself. A second, shorter cast shadow appears on the boy’s upper thigh, matching the direction within a few degrees and demonstrating that the vector remains stable on a curved receiver. Finally, a form shadow along the crown’s crease aligns with the cast shadows. Each cue samples the illumination field from a different geometric context—a flat background, a curved body surface, and a continuous form shadow—and yet all converge within a narrow angular window. If the brim line were misinterpreted, these cues would diverge sharply, since their dependence on surface orientation and occlusion geometry is independent.

• (Bottom-right rock) Dalleur interprets a dark patch at the base of a rock in the bottom-right corner of the frame as a cast shadow from the rock’s ridge onto the adjacent ground. Mark objects that the shadow is short and indistinct, comparing it to a lapel feature on a man’s jacket that he regards as analogous. The comparison, however, fails: the lapel mark is a form shadow, a gradient caused by curved fabric turning away from the light, not a cast shadow projecting from one surface to another, and therefore it is not a valid cue to define a beam line. By contrast, the rock shadow presents a crisp outer boundary where light meets shade across a nearly planar receiver. Though its visible extent is limited, it covers a long enough baseline on the original plate that any plausible tracing error wouldn’t change its orientation by enough to make a meaningful difference.

Dalleur’s argument does not require that he obtain a precise value for the angular elevation of LSa. It requires only that the illumination vector be constrained to a range that does not include the solar elevation. Even when tracing uncertainties are taken into account, the hat and rock cues are sufficient to accomplish that. All three beam lines intersect at healthy angles; that orthogonality makes the vanishing point geometrically stable: small shifts in any line displace the intersection minimally. Moreover, the projective cues are validated by non-projective cues. Here are examples:

• (Shawl folds) A curved transition between light and shadow is visible along the girl’s shawl, where the fabric folds smoothly across her shoulder. At the point where this boundary is locally continuous and monotonic, Dalleur draws a tangent to the curve. Although this method is less informative than those based on object–shadow pairs, it provides an independent corroboration of the azimuthal component of the illumination. Since azimuth is geometrically coupled to elevation, that provides strong evidence for a low elevation solution.

• (Collar fold) A sharply defined transition between light and shadow runs diagonally along the collar fold of a boy’s jacket, where the fabric bends toward the camera. Because the collar fold is stiff, planar, and geometrically well-defined, the tangent direction here is even more reliable than that derived from the soft, irregular curvature of the shawl: the underlying structure is simpler, and the boundary is sharper and less affected by local texture or drape. That confirms that a low elevation solution is consistent across surface orientations.

• (Pant knob) There are symmetrical reflections on a perpendicular knob on the boy with the hat’s pant leg. On a vertical, cylindrical surface, a specular highlight appears where the surface faces halfway (horizontally) between the camera and the light. The highlight marks the locus where the angle of incidence equals the angle of reflection. Because the highlights are near the front, that forces relative azimuth to be modest. A high-elevation solution requires an extreme azimuth, but that would have displaced the highlights and diminished their intensities.

3.1 - Face directionality?

Mark writes “most of the people in photo D115 don’t seem to be facing a relative azimuth of ~30˚”

Face directionality is an unreliable indicator of relative azimuth. People can rotate their eyes ~45–55° horizontally without moving the head.¹⁵ Soft focus and shallow depth of field smear the landmarks needed for pose estimation, so small pixel shifts translate into large errors. Not to mention that, according to witnesses at the Cova, the gazed source was in motion and there were multiple distractors. ‘Which way the faces seem to point’ should never take precedence over physical lighting cues.

Mark admits that his analysis of face directionality yielded inconsistent results. He claims that the four individuals in the center of the frame appear to face 60–80° to camera-right (based on a Blender simulation with FOV = 62°). One page later, he begrudgingly admits that, given the same assumptions, it is “somewhat less plausible” the two figures on the right are facing ≈70°. If Mark’s paradigm were valid, the yaws would have converged; instead his own runs suggested incompatible headings.

3.2 - Left-most boy?

Mark writes “But to me, by far the most powerful argument for a high LSa relative azimuth comes from the shading of… the leftmost boy’s right leg… There are a few other shadows and highlights in D115 that seem difficult for Dalleur’s model to explain, but this one seems to me the most telling.”

Mark has updated his document to agknowledge that “Prof. Dalleur suggested by email that the darkened leg here is in fact the shadow of the photographer (and his equipment)…” Dalleur provided a Blender simulation as a proof of concept:

4. Rapid Drying

Scott acknowledges that his theory “can’t explain why some people said that they were sopping wet when the miracle began, but their clothes had dried completely by the end of it... It must dismiss this as people not being very good at assessing exactly how wet their clothes were, or how quickly one would normally expect them to dry.”

The consensus of witnesses was that there had been continuous precipitation and heavy overcast at the Cova until around the time that the Miracle of the Sun occurred. That is corroborated by meterological data from civil and military observatories. It is also corroborated by photographs that were taken before and during the event:¹⁶

If the crowd had been standing in the Mojave Desert on a sunny and hot day, it would have taken hours, not minutes, for saturated garments to fully dry. The ambient temperature and humidity at the Cova would have been profoundly unfavorable to non-negligible drying at the time of the event. The temperature on the Fatima plateau at midday in mid-October ranges from 15–20 °C. Even at the upper end, that is mild. Relative humidity after prolonged rain would have been near 100 %. In such an environment, the driving force for evaporation would have been nearly zero.¹⁷

Yet a large number of witnesses reported that, during the Miracle of the Sun, their garments went from being sopping wet to being completely dry. For example:¹⁸

Mr. Louis Lopes… went to Fatima and told us afterward that although he had taken every precaution to avoid the rain (because he suffered from bronchitis) he was soaked through from his feet to his waist. But as the sun came down from the sky he noticed, to his astonishment, that he was completely dry.

That is corroborated by the dry clothes in a photograph taken after the event:

4.1 - Infrared shadow

The most conclusive evidence of miraculous drying is an infrared shadow in D115. Dalleur observed that there are three distinct tonal regions on a boy’s pants: a dark band across the hips matching where his buttoned jacket would have fallen, a vertical darker line along the fly corresponding to moisture retention in the placket, and a lobed patch toward the left hip consistent with partial occlusion—likely from an arm or jacket fold. Dalleur argues that these are wet–dry boundaries produced by a sudden burst of infrared radiation that dried exposed surfaces while leaving shielded areas damp. The geometry of these boundaries, their crispness, and their alignment to occluders make them direct physical evidence of selective, instantaneous drying.

Scott defers to Mark Grant’s rebuttal to Dalleur’s interpretation of the tonal regions. This time, Mark’s argumentation is pretty far-fetched. His alternative hypothesis is that both S and W are unknown photographic defects. That is extremely unlikely:

• (Scene-locked geometry) The boundaries of S and W track the three-dimensional clothing geometry, not the plate or image borders. The main dark swath (S) runs exactly parallel to the jacket hem and stops at seams; the secondary region (W) bends along a natural seam and follows the curvature of the thigh. Downward, it fades gradually along vertical corrugations and halts at the knee crease. Photochemical or processing artifacts never respect subject geometry so precisely; they cut across folds and edges indifferently. This ‘geometry-locking’ shows that the contrast belongs to the scene itself, not the development process.

• (Optical signature of wet cloth) The tonal and textural qualities of S and W match, almost textbook-perfectly, what wet wool or cotton looks like in black-and-white photography. Darker, lower-contrast, and slightly smoother surfaces due to water filling air gaps in the weave (reducing internal scattering). No loss of detail or resolution, meaning the darkness arises from absorption, not from chemical fog or under-exposure. The contrast ratio—roughly one to two stops difference from the adjacent zone—is exactly what’s expected from saturated vs. dry fabric.

Mark tries to motivate his hypothesis by pointing out that “there are weird discolorations in other photos that aren’t well explained as IR shadows, but more likely some unknown physical defect somewhere in the photographic development pipeline. Thus, it may be plausible that [the tonal regions] represent a similar kind of unknown defect, too, whose nature we don’t need to explicitly diagnose.” That strategy is paradoxical. If the ‘weird discolorations’ don’t have the features that make it plausible to interpret the tonal regions as IR shadows, then there isn’t parity. Additionally, the assumption that ‘anomaly’ implies ‘photographic defect’ is fraught and question-begging. However, let’s take a look at the first anomaly that he identifies:

The green-circled region in D114 is best explained as a wet–dry contrast. Mark is right that it cannot plausibly be the imprint of an occluder. Rather, the contrast likely formed after the drying event, when the man placed his knee back down on the wet soil, re-wetting only the portions of the fabric that touched the ground. When the man placed his knee back onto the wet soil, the fabric at the point of contact rapidly absorbed water along the path of greatest capillary conductivity, which in wool trousers is the stitched side seam. This produces a linear, high-contrast front between the saturated and still-dry zones. The fact that the boundary aligns with the seam makes it very unlikely to be a photographic defect. Far from weakening the case, the D114 knee provides positive, physical evidence for rapid, complete drying. It also demonstrates the folly of Mark’s readiness to assume that anomalies are defects.

Finally, Mark objects to Dalleur’s interpretation of the tonal regions. His main argument is that S is morphologically inconsistent with an occluder imprint:

Mark concedes the magenta band is a very hard, roughly linear edge that is “parallel to the ground” and “kind of looks like the edge of a jacket.” However, he objects that the blue darkness down the fly is “as dark as anything in the magenta region,” yet not an imprint of a separate object. That is correct—and it fits the drying model: after exposed surfaces dried, residual water was trapped inside the occluded jacket region. That water subsequently drained downward under gravity through the seam allowance and inner layers, following the natural capillary channels of the garment. The fly placket, composed of multilayered, densely stitched fabric, then retained that draining moisture because of its strong capillary continuity and limited ventilation.

Mark calls the green area “elbow‑shaped,” diffuse on one side, and therefore puzzling for a jacket hem imprint. On the selective‑drying account, the green zone is not a second hem; it is a penumbral occlusion formed by the boy’s forearm/hand/stick and the left jacket panel hovering at small, varying distances from the cloth. The curved surface of the hip and irregular shapes of the arm and coat front would naturally project an elbow-like lobe, while moisture migration along seams after exposure would further soften and elongate the boundary. In other words, the green patch’s morphology—rounded, fading on one side, anchored to garment structure—is exactly what you would expect from partial shielding under directional radiant drying.

5. Replies to Objections

5.1 - How did it end?

Scott writes “Dalleur claims the light source was not the sun at all, but some sort of artificial miraculous object. But if this were true, how did the miracle end? No witness describes seeing the pale sun disappear. They only say it went back to its usual place in the sky. Later in the day, the clouds cleared and it became a normal sunny day. But nobody reports seeing two suns. At some point, either the first light source must have vanished (which would have been noticed), or there must be two suns in the same sky (which would also have been noticed).”

Actually, several witnesses implied there were two suns. Raio de Luz reported “Suddenly, a luminous disk the size of a large host, but which everyone could gaze upon as one gazes upon the moon, appeared, as if it had detached itself from the sun (or rather, placed itself before it)…” Maria Campos reported “We saw a silvery veil, rounded in shape, as if it were a full moon… Cries were heard from all sides as it emerged from the sun like a white, shining, snow-like shape… coming toward us, returning to the sun again, and finally hiding for the third time among the clouds.”

There is a subtle detail in the testimony from a witness in Alburitel that sheds light on how the miracle ended. Inacio Lourenço reported that “After ten minutes, the sun resumed its place, pale and without splendor.” That is at variance with the witnesses at the Cova that reported that the Sun ‘regained its brightness.’ That discrepancy corresponds to a geometric prediction of Dalleur’s model. Scott is right that witnesses reported that, at the end of the miracle, the ‘Sun’ returned to its normal position. If they were at the Cova, it appeared to regain its brightness. If they were at Alburitel, it didn’t. LSa was aligned with the solar azimuth relative to the Cova, but not relative to Alburitel. From the perspective of the Cova, as soon as the Sun was unoccluded, its overpowering brilliance absorbed the dimmer LSa. From the perspective of Alburitel, the geometry was offset, so the Sun couldn’t absorb LSa at the end of the event.

On the other hand, if the Miracle of the Sun were subjective, there is no discernible reason why diverse crowds would have experienced a synchronized conclusion. The duration of perceptual effects is sensitive to gaze direction, blink rate, pupil dilation, cloud cover relative to one’s exact position, and individual variability in retinal chemistry. If the event were nothing more than a collective illusion, some would have stopped perceiving motion seconds or minutes earlier than others, while others might have continued to perceive oscillations or chromatic shifts long after the rest had ceased. Yet witnesses agree that the event ended suddenly and simultaneously.

5.2 - Everywhere else?

Scott writes “although Dalleur’s theory somewhat makes sense for Fatima, it stumbles for Ghiaie and becomes completely incoherent for Benin City. At Ghiaie, the miracle was seen 15-25 miles away to the east (in Tavernola), but not 15-25 miles away to the southwest (in Milan), even though the line-of-sight from Milan was clearer. In Benin City, the miracle was localized entirely to one large field, while the rest of the city (population 1.5 million) saw nothing.”

This objection rests on the unmotivated assumption that Fatima, Benin City, and Ghiaie had to have the same physical implementation and intended scale. Miracle believers have no reason to accept that. In the case of Benin City, it is clear from the context in which it occurred that, if it was a genuine miracle, it was merely intended to commemorate Fatima, not replicate it. In the case of Ghiaie, it is entirely possible the position or directionality of the light source was fine-tuned to be visible to the east but not the west. Scott gripes that “This would contradict Dalleur’s Fatima analysis, since one of the most dramatic testimonies comes from the city of Minde, which is on the opposite side of the presumed light source from Fatima.” So what? It is plausible that the author intended to limit the scale of Ghiaie compared to Fatima.

Also, Scott’s assertion that Milan would have had a clearer line of sight to an anomalous light source near Ghiaie than Tavernola is dubious at best. First, it sensitively depends on the position and elevation of the source because the Milan and Tavernola horizons are set by very different occlusion geometries. Tavernola’s effective horizon toward Bonate runs across a lake-to-plain corridor. There is no continuous, close ridge on the Tavernola to Ghiaie azimuth playing the role that Monte Canto plays for Milan. Second, the optical environment of the Po Valley make detection from Milan harder. There is chronic fog and haze in the lower Po Plain.¹⁹

5.3 - Domingos Pinto Coelho?

Scott writes “I was immensely heartened when I finally found the primary source for one of the classic Fatima testimonies - that of the lawyer, Catholic activist, and Portuguese senator Domingos Pinto Coelho. After discussing his awe at witnessing the miracle - the part everyone always quotes in their Fatima write-ups - he said (using the royal “we” for an official newspaper column): ‘Was what we had seen in the sun something exceptional? Or could it be reproduced in similar circumstances? This very analogy of circumstances was provided for us yesterday. We could see the sun half-obscured by clouds, as on [October 13]. And honestly: we saw the same successions of colors, the same rotation, etc.’”

The Portugese historian Costa Brochado cast doubt on the integrity of this report:²⁰

The articles of Dr. Pinto Coelho in A Ordem seem to be very much exaggerated. It is difficult to understand the manifest confusion he establishes between the phenomena at Fatima… and the alterations in solar light that he says he saw in Lisbon some days afterward. But in any case the historical value of the articles of the leading Catholic organ is almost nil… We believe that we can declare, after patient research on the matter, that the articles of Dr. Pinto Coelho ought to be read from a political point of view, since their objective was, as the author himself came to declare, to serve as the devil’s advocate…

It is also worth noting that a witness to a similar event disagreed with Coelho:²¹

The Bishop of Portalegre and Mrs. Maria de Jesus Raposo relate that being with others in Torres Novas, on October 20th… they saw the rotational movement of the sun and the change of colors. The same lady states that these manifestations in the sun were quite different from those in Fatima and did not have the importance of those of October 13th past. It is of the utmost importance to know what these differences are for she was present at both.

1. https://slatestarcodex.com/Stuff/fatima_statements.odt

2. Amorim, Diogo Pacheco (1960). “O Fenómeno Solar de 13 de Outubro de 1917.” O Instituto, vol. 122, pp. 145–204. Coimbra: Imprensa da Universidade de Coimbra.

3. Ibid.

4. De Marchi, John, I.M.C. The True Story of Fatima: A Complete Account of the Fatima Apparitions. Catechetical Guild Educational Society, 1956 (rev. ed.).

5. Haffert, John (2006). Meet the Witnesses: Of the Miracle of the Sun. American Society for the Defense of Tradition, Family, and Property–TFP.

6. Ibid.

7. Brochado, Costa. Fátima in the Light of History. Milwaukee: Bruce Publishing Co., 1955.

8. Ibid.

9. Documentação crítica de Fátima: seleção de documentos (1917–1930) (L. C. Cristino & A. T. da S. Silva Neto, Eds.). (1999). Fátima: Santuário de Fátima, Documento 133.

10. Jaki, Stanley L (1999). God and the Sun at Fatima. Real View Books.

11. Amorim (1960).

12. Ibid.

13. Ibid.

14. Instituto Nacional de Estatística (Portugal). Recenseamento Geral da População de 1911. Lisboa: Instituto Nacional de Estatística, 1914.

15. Lee, W. J., Kim, J. H., Shin, Y.-U., Hwang, S., et al. (2019). Differences in eye movement range based on age and gaze direction (Eye, London, 33(1 Pt 2):1). PMC.

16. Dalleur, P. (2021). Fatima pictures and testimonials: In-depth analysis. Scientia et Fides.

17. Gluesenkamp, K. R. (2021). Efficiency limits of evaporative fabric drying methods. Drying Technology, 39(17), 2421–2430.

18. Haffert (2006).

19. NASA Earth Observatory (2003). Haze in the Po River Valley, Italy.

20. Brochado (1955).

21. Jaki (1999).