Where were they? In lots of places, according to both scientific / archaeological and written evidence, but you are right in saying that Wikipedia gives the impression that wildfires and forest fires didn't happen for much of human history (see, for example, List of wildfires - nothing pre-19th century).
This impression is entirely false, but wildfires occurring in unpopulated areas or where people were not literate would not have been recorded in the way that they can easily done in more recent times. Further, fire was often used to clear land. Thus, in ancient and medieval Greece for example,
Fire became an essential component of the landscape.
In other words, fires outside of cities were probably not deemed significant enough to be recorded. Even when wildfires were recorded, there is little indication as to the size of the area affected; there were no government or environmental agencies paying investigators / researchers / scientists to do it.
Nonetheless, there are written accounts of wildfires (as well as scientific evidence) from the ancient, medieval and early modern periods.
In antiquity, Vitruvius, Thucydides, Theophrastus and others all made references to spontaneous wildfires. For example, Thucydides writes when referring to a fire made by the Peloponnesians during an assault on Plataea that
The consequence was a fire greater than any one had ever yet seen
produced by human agency, though it could not of course be compared to
the spontaneous conflagrations sometimes known to occur through the
wind rubbing the branches of a mountain forest together.
Thucydides also mentions a fire started by a soldier which got out of control due to the wind. It seems that most references to wildfires in ancient sources do not mention specific incidents, just the fact that they happened. Homer, in the Iliad, uses a wildfire to describe Achilles' rage:
As a fire raging in some mountain glen after long drought - and the
dense forest is in a blaze, while the wind carries great tongues of
fire in every direction - even so furiously did Achilles rage...
and again when describing the armies charging into battle. The Roman poet and philosopher, Lucretius (died c.55 BC) mentions in De Rerum Natura that some forest fires were caused by lightning. Though the fire itself is described as the "flamy heat with awful crack and roar", he then describes a positive end result:
The forest trees and baked the earth with fire, / Then from the boiling
veins began to ooze / O rivulets of silver and of gold, / Of lead and
copper too, collecting soon / Into the hollow places of the ground.
For the most part, though, ancient writers focused on deforestation caused by human actions, and the consequences thereof (e.g. lack of timber for shipbuilding).
In Spain, the article A comparison of the medieval and the current fire regimes in managed pine forests of Catalonia mentions that there are records of fires in the forests in the mountains near the city of Tortosa. The authors looked at the period 1370 to 1466 and found that
Throughout 1379-1466, the mean number of fires per year in the Port de
Tortosa area was 0.62 (S.E.0.09), with two peaks, one in the
1400-1409 decade and the other in the 1420-1429 decade...
They also found that
Most fires currently occur in summer, and the same pattern is observed
during 1370-1462 period.... These fires also account for most of the
burned area or the suppression effort, respectively. In the Middle
Ages period, however, autumn and winter fires were more abundant than
The Russian Niconovsky Chronicle records huge fires for the years 1365 and 1371 when dark spots on the sun were also observed. The chronicle relates:
There were dark spots on the sun, as if nails were driven into it, and
the murkiness was so great that it was impossible to see anything for
more than seven feet. . . . Woods and forests were burning and the dry
marshes began to burn and the earth itself burned, and great fright
and terror spread among men.
In England's Sherwood Forest there was a fire in 1624:
‘In 1624 during the great drought of that year Sherwood Forest
suffered a great fire, and White (Worksop, the Dukery, and Sherwood
forest) quotes from a manuscript preserved in the British museum in
which it says that there was such a mist of smoke and particles that
people thought it was an eclipse of the sun, but when the true cause
was discovered ‘there came command from the justices to raise the
country there about and bring pickaxes, spades and shovels to make
dikes and trenches to break the fire in the forest’. This fire, four
miles long and half a mile wide, was stopped at the wood between
Mansfield and Nottingham’. (Illingworth Butler 1946)
There is also ample evidence of fires elsewhere in the world. Stephen J. Pyne has written several books on fires, and covers the pre-European contact period in North America as well as the early European settler period. Also for North America, there is the article Medieval warming initiated exceptionally large wildfire outbreaks in the Rocky Mountains. Researchers found that between 800 and 1300, the frequency of fires increased and 83% of trees burned. One of the authors, Gonzalo Jiménez-Moreno of the UGR´s Department of Stratigraphy and Palaeontology explained that:
we have proved that the frequency of wildfires increases when
temperatures increase, something which occurred during the Medieval
Warm Period and which is also occurring at present.
Evidence from the Blue Mountains in Oregon and Washington shows that wildfires were actually more common before 1900:
After 1900 fires became extremely rare, indicating a change in
controlling factors as livestock grazing and fire suppression became
One problem, though, is separating fires caused by human activity from naturally occurring ones. For example, in Bohemia and Moravia there are over 30 "place names of the type Zdar (burnt land)" dating from the 1230s but it is impossible to say if these relate to fires caused by human activity or if the occurred naturally.
It is evident, though, that human activity was significant:
Techniques have been developed for dating the year of fire occurrence
and, to a lesser degree, for estimating the severity and extent of
historic fires as well, thereby extending considerably the fire
history for an area. By far, the overriding theme emerging from fire
scar studies affirms the drop-off in fire frequency that accompanied
Euro-American settlement of the North American continent, with the
biggest reductions in lower-elevation forests. Similar dropoffs have
been noted in Australia and elsewhere.
Source: Philip Omi, 'Forest Fires: a reference handbook' (2005)
Nonetheless, naturally occurring fires were noted in California:
one ethnographic report describes a massive wildfire in San Diego
County prior to the time of European contact that resulted in a
significant migration of Native American residents to the desert.
There is also, unsurprisingly, the problem of evidence that has disappeared over time:
Evidence for prehistoric fires has all but disappeared from European
landscapes, and we can imagine that much of the global history of fire
over the millennia remains undisclosed.
In comments below, Pieter Geerkens makes the very valid point of distinguishing "between crown fires and others" and that it is only the (destructive) former which are likely to have been recorded. Undergrowth fires, on the other hand, are "part of the natural cycle" and Mr. Geerkens adds
Crown fires are much less frequent in the absence of active human
suppression of undergrowth fires. It takes very high temperatures -
thus an unusual undergrowth accumulation - to light bark and start a
Not all destructive fires are crown fires, though. The term Stand-replacing fire is used for a fire that "kills most or all of the trees in the stand" (which includes crown fires). The article Global trends in wildfire and its impacts: perceptions versus realities in a changing world by Stefan H. Doerr and Cristina Santín has more on this.