(Starting note to Readers : This is a very long article. Part way through this I said to a friend, "I've got to get a hobby". His reply was "Dude! This is your hobby".)
“
Medieval Wisdom explores the ingenious solutions of the Middle
Ages that still amaze us today. Dive into a past filled with
creativity and discover how these ideas can inspire the present.”
https://www.youtube.com/@MedievalWisdom
I was made aware of
this collection recently, related to a specific title : “What
Was Bog Iron? The Medieval Blacksmith's Rustproof Metal You've Never
Heard Of” that was made available on February 9 2026.
Part One : ' the
Medium is the Message'
Medieval Wisdom
appears to have no associated web site, a simple internet search did
not find anything beyond the YouTube channel. This channel is given
as dating from 2011, but the oldest content only dates from the past
year (January 25, 2025), with a total of 181 videos listed. The
location given is USA, but there is no information of just who is
behind the collection.
The same individual is
shown in the most recent videos, who, judging from appearance, is an
Early Medieval (Viking Age?) re-enactor. This person (presumed the
producer?) is never identified. Typical offerings run about 20
minutes. Worth noting here that this is a creation rate of about one
every three days. For research, writing, gathering visuals, narration
and final editing?
The titles of almost
all these videos contains phrases like 'Copy This...',
'...That Scientists Now Say Were GENIUS', '...That Modern
Engineers Can't Replicate', 'BANNED...'. The thrust of
almost all the content is 'Medieval methods are better, They were
smarter than us'. Pure click bait titles, deliberate selection of
evidence to stress bias.
As soon as you launch
any of the videos several things become apparent :
All narration is
via a computer synthesized voice, AI generated from a prepared text.
This makes the pacing irregular, especially in terms of incorrect
stress and flow. Any non-English names or terms are most commonly
incorrectly pronounced for just this reason. Quite annoying for
content based on the European Medieval period, country locations,
and the many languages involved.
Video content is a
combination of AI images, sequences from video games, very short
clips from feature films, images and clips taken from personal web
sites. Likely all used without any permissions, with no specific
credits given. Clips are uniformly very short. The use of AI
generated segments specifically presents a problem for content
presented as historical facts.
Perhaps because of
the sources used, often the visuals do not match the actual
narration, sometimes found to be not at all representing what as
being described.
Many of the points
presented are at the least distorted, if not misrepresented. (based
on samples viewed)
I viewed portions of
six different of these videos. Four picked more or less at random
(one the first in the series). Two were topics in which I could
certainly be considered 'expert' in (see below).
The narration is
cloaked in the tones of authority. However, at least based on the
samples viewed, information is specifically framed to stress
'Medieval was Smarter'. This represents a clear bias, often outright
distortion, with some points dramatically stressed, while related
aspects missing entirely.
The tone of the
productions, and the visuals themselves, added to the extremely short
creation schedule, all suggest these videos are content primarily
generated by AI systems.
Part Two : 'What was
Bog Iron?'
'What
Was Bog Iron? The Medieval Blacksmith's Rustproof Metal You've Never
Heard Of'
https://www.youtube.com/watch?v=pe3CgMzwVD0&t=4s
(text description)
“Medieval blacksmiths harvested iron from
bogs and swamps — a renewable resource that regenerated every
twenty years. The resulting metal, naturally rich in phosphorus and
silicates, resisted rust far better than modern steel. From Viking
longship nails that survived a thousand years to wrought iron
railings still standing in New Jersey, bog iron challenges the
assumption that modern metallurgy is always superior. This video
covers the complete process: how bog iron forms, how it was smelted
in bloomery furnaces, why phosphorus gives it corrosion resistance,
and what we lost when industrial blast furnaces replaced the medieval
method.”
Within
two days, this video was posted to the 'Iron Smelters of the
World' discussion group on Facebook, which consists of the
majority of the participants in 'Early Iron' activities, admittedly
dominated by Europeans and North Americans. (I have been an active
core member since inception.)
The
very first set of comments within the group related to the repeated
inclusion of both images, and often extremely short video clips, of
well known current iron makers. As it turned out, none of whom
even knew about this production, or certainly had not given any
permission for the use of their own copyrighted images or material.
This was material gathered from the open internet.
A
large number of things jumped out at me on viewing this specific
video.
Although
a list of references is given in the written description, these are
clearly the result of a simple internet search. No direct links are
given. On examination, these are certainly not the best easily
available sources for this topic. Many of these sources are dated,
even questionable, in terms of the information they presented. This
all becomes an even bigger problem when you consider the sources
where so many of the images and short clips were gathered, which
commonly have far more complete and accurate information, were not
referenced (and I question even consulted).
So
perhaps not unexpected that the presentation itself would not be the
most reliable?
Looking
at a very fast internet search, it would appear the primary
source used for for any information presented related to the history
of iron smelting itself was actually taken from the Wikipedia article
'Bloomery' – this includes a number of the images used. (1)
(from
start) “ You are looking at mud. Foul smelling, sticky mud from
the bottom of a Scandinavian peat bog... because inside that mud is
iron.”
A
major problem through this presentation is that there will be no
distinction between fresh 'primary bog iron ore', as is shown, and
rock like limonite, which is a much older (often ancient) formation
from the same initial source. This is perhaps not surprising, as 'bog
ore' is a very loose term, often applied locally to any iron ores
found in proximity to water.
(starting
about 0:40)
The
short section on 'corrosion of mild steel' is distorted at best.
-
First off, The juxtaposition of the endurance of railings made of
antique wrought iron, likely referring the Trenton Iron works which
was established in 1854 (2),
against economics of modern industrial mild steel infrastructure is
deliberately misleading.
-
Second, the section of rusted to failure modern pipe shown at
roughly 1:00 is most likely cast
iron, the way it breaks on hammer blows certainly suggesting this.
Narration stating “... a problem that Medieval blacksmiths
partially solved with swamp mud” is both sensational and
extremely simplistic.
(about
3:00)
In
the discussion of the correctly described 'renewable' nature of
continuously deposited bog iron ore in undisturbed
wetlands, another extremely important factor is left out. Although a
fresh layer of the ore can re-accumulate in as little as 20 years,
the total amounts are relatively small. In practical terms, this
amount is typically 'another smelt's worth', so in the range of 10 –
20 kg at best. The expected yield from a single smelt in this size
would be raw bloom in the range of 2- 5 kg. Stating, “economically
useful” is a distortion, which
could only apply to the smallest Early Medieval unit, the extended
family farmstead. (3)
(This reality needs to be born in
mind when considering the description of potential yields later in
the presentation.)
(about
3:30) “In Scandinavia and Finland, bog iron dominated iron
production from roughly 500 to 1300 CE. That is eight centuries of
continuous use.”
Archaeology
places the start of the Iron Age in Scandinavia to roughly 500 CE,
and bog ores would be the main type. Bog ores were still being
utilized in Sweden up into the 19th century. This is
considerably longer than the eight centuries stated in the
presentation. (4)
(at
3:45)
The
narration is about iron smelting in Iceland. The initial image shown
is in fact a reconstruction experiment at the Ribe Viking Centre, in
Denmark. (5) Pictured is a
clay built furnace, most certainly not the type used in clay
poor Iceland. (This is actually described later in the program.)
(about
4:05) “The process of
turning bog mud into usable iron followed a sequence that was
fundamentally the same over all of these regions, even though they
had no contact with one another. The universality of the method
suggests it was not invented once and spread, it was invented
multiple times, independently, because the logic of the process is
self evident once you understand the material.”
A
number of core problems with this statement.
It
is incorrect that there was no contact between these Northern
European areas. Since the stress of this presentation is celarly on
Scandinavia, and Norse expansion over the North Atlantic, these
people shared a similar material and technological culture, even
related languages. Artefacts prove that there was considerable and
regular trade within these regions and well beyond into England,
mainland Europe, the Middle East and even beyond.
The
dating of iron smelting sites from the Middle East into Europe
indicates the earliest smelted iron is from norther Iran / Iraq (at
roughly 1500 – 1200 BCE). The iron smelting process spreads
relatively quickly west and north through Europe over the next 700
years (as mentioned above)
“...the
logic of the process is self evident once you understand the
material.” This is a massive overstatement (as someone
personally involved in the process of recovering these exact same
skills). The initial concept that red mud could possibly be
converted into useful metal is completely counter intuitive.
Individual bog ores, at specific locations, can can have quite
different elemental concentrations, no all bog ores give good
results. An iron smelting furnace operates at higher temperatures
than other furnaces. (Hundreds of degrees Celsius beyond that of a
pottery kiln for example.) A single smelt consumes a large amount of
carefully prepared resources, and success is dependant also on
following the correct working sequence over many hours. (6)
If
nothing else, the archaeology shows that a basic concept of how an
iron smelting furnace may function has been repeatedly adjusted by
widely separated workers to suit specific locally available
resources. (Reference to this, related to Iceland, is featured later
in the video itself.)
(starting
at 5:10)
Roasting
bog ores? There are a number of mis-conceptions given in this
section:
Baking
dry? Although it is true that water can consume considerable energy
in a burning system, the simple truth is that so much energy is
available in a smelting furnace that moisture has little practical
effect. Merely increasing the effective stack height of the furnace
will quickly expel any remaining dampness in the ore.
Porosity?
One of the huge advantages of smelting bog ores is that they are
already quite porous and additionally quite easy to break up into
suitable sized small pieces for adding to the furnace. (It needs to
be remembered that the reduction process that will be described
later in the production is a gas on solids one.)
(starting
at 5:40)
Furnace construction? The primary distortion here
is that the term 'Medieval' is being applied so loosely here as to be
misleading. The furnaces described are from both from early history
and very regionally specific. Furnaces built under the Romans, tend
to be much larger (~ 1.5 m ID / ~ 2m tall) that operate under passive
draw air. The introduction of water powered machinery into Europe (so
very roughly 800 – 1100 CE, starting in the south and defusing
northwards) will significantly change the scale of furnaces. What is
described are furnaces most typical of the post Roman period, from
roughly 400 – 1000 CE in Europe (often called 'the Dark Ages').
Although clays, and especially clay mixtures, are ideal construction
material for smaller, limited use furnaces, stone construction was
also commonly used for more 'industrial' multiple use structures.
(starting
at 5:58) “The furnace was loaded with alternating layers of
charcoal and crushed, roasted ore. With a typical ratio of one to one
by weight adding seven to fifteen kilograms of ore incrementally over
two to four hours”
The
way this is stated represents two entirely different methods of
working sequence during a smelt :
“The
furnace was loaded with alternating layers of charcoal and crushed,
roasted ore.”
This
is a method that only will work properly with massively tall,
typically passive draft iron smelting furnaces. The primary example
are the types documented and used in Africa, and this method also
works for the earlier Roman builds. Although initially placed in
thick alternating layers of ore and charcoal, in effect the tumbling
of individual pieces of charcoal within the furnace as they fall
will evenly mix ore and charcoal together. This is critical, as a
thick slab layer of ore will effectively choke the furnace
otherwise. Although this method is still incorrectly illustrated for
the small furnaces featured in this presentation, it has repeatedly
been proven not to correctly function at this scale.
“With
a typical ratio of one to one by weight...”
Not
at all, this is incorrect on two factors (or at least indicates very
poor method).
As
the reduction process starts, it is exothermic, this
additional energy will raise temperatures inside the furnace. This
excess temperature is controlled by increasing ore amounts combined
with fixed measures of charcoal. This “one to one by weight” is
only an averaged amount, determined mathematically afterwards from
the overall totals used. This is an equivalence based on simplified
reports by only a small number of makers. It does not take into
account the total amount of charcoal to pre-heat then bring the
furnace to correct operating temperature, or during the final burn
down phase. (7)
“...adding
seven to fifteen kilograms...”
This
blanket statement is so simplistic as to be mis-information.
a)
Later in the program there will be a (greatly simplified) discussion
of the dynamics of slag bowl formation for the collection of a
workable iron bloom. There needs to be a certain amount of ore
material added to establish this system, before any bloom iron at
all will be created. The elemental components of a given ore
determine this formation, so certainly there can be no standard
given as a minimum ore amount. (Our own experience, working with
ores in the 55 % elemental iron content range, is that a minimum of
8 kg ore is required before any recoverable bloom will begin to
form) (8)
b)
Once the working slag bowl system is formed, the more ore that is
added, the larger the bloom. Importantly, the bloom compresses
itself as more and more weight is added. Although it is true some
iron may be recovered, utilizing extremely high elemental iron
content bog ores with total additions in the range of 10 + kg, the
resulting blooms will be lacy, fragmented and so extremely difficult
to effectively forge into workable iron bars (9)
Larger ore amounts (using the same ores) have been shown repeatedly
to greatly improve not only overall yield percentages, but also
greatly increase bloom density itself. (10)
“...adding
...incrementally...”
Although
the functionally the ideal method, this directly contradicts the
description made in the previous sentence. (??)
(at
6:15) “The smelting temperature reached between 1100 to 1300
degrees celsius.This is critical to understand.”
And
also wrong. A bloomery furnace is operating at temperatures, recorded
at the hot zone above the tuyere, in the range of 1300 to 1450° C.
Obviously the very base of the furnace (below the slag bowl) will be
cooler. These temperatures are well documented scientifically and
from measurements by any number of experimental teams. (This
including a team specifically featured in this presentation.)
(from
7:00) “ A bloomery furnace typically only converted 10 to 20
percent of the iron that was in the ore into usable bloom.”
Importantly,
this statement does not clearly distinguish if what is meant is the
iron ore weight or the elemental iron amount. As discussed above, the
exact elemental iron content of the staring ore has an impact here,
as does the total volume of ore used. Again, this is not the evidence
from archaeological furnaces, or the experience of modern bloomery
iron makers. Modern makers typically expect a minimum of 20 % return
by weight of ore into a bloom, a 10% return would be considered a
failed effort. Ranges into 30 % are common, and for large volume ore
smelts results in the 40 + % range have been achieved) Considering
elemental iron content to proportion converted into extracted blooms,
my own team average is 36.6 % recovered. (11)
(at
7:35) “The result was wrought iron, a low carbon iron that
could easily be shaped into tools, nails, hinges, knives and
weapons.”
A
complete misunderstanding and gross simplification:
In
actual fact, carbon alloying content will vary between individual
blooms. Even within a single bloom, the carbon content will vary
between bottom (first deposited) an top (last deposited and
typically at a higher temperature).
An
experienced smith can estimate the relative 'hardness' of metal
being hammered by how easily the piece will deform under the same
hammer blows. This harder metal is more suitable for objects
requiring physical hardness, tool working surfaces and knives being
the obvious applications. Yes, softer iron requires less effort to
forge and is easier to fire weld. This quality was most likely the
more desired product, as it would be used for the bulk of every day
items and hardware.
Weapons,
however, require qualities that are at opposition to each other. A
very hard carbon iron alloy would certainly take, and hold, a sharp
cutting edge, but at the same time hardness increases, so does
brittleness. A softer iron would accept shock well, but would be
most likely to simply bend in use, and would never hold a sharp
edge.
As
stated, slight variations in carbon content would be the primary
reason for these variations in hardness. This is not the only reason
for variation in the iron metal made from bog ores from different
locations. There can be other metallic trace elements present, which
can have impact on the qualities of the resulting blooms.
(at
8:18) “ But bog iron ore is naturally rich in phosphorus.”
Distortion.
This is absolutely dependant on the specific ore sample, certainly in
location and also over time. This concept of different deposits
resulting in differing qualities in the smelted iron was well known.
(As an ongoing natural accumulation, changes in local environment
over time can alter the characteristics of deposits, even at the same
physical location.) The bog ore recovered in the archaeology at
L'Anse aux Meadows, described in detail later, had almost no
phosphorous content as a good example.
(from
9:25)
The right hand image is labelled 'Ancient bog iron,
pristine glassy surface'
To
my (experienced) eye, this looks to be a piece of waste slag.
The smooth, dark black, glassy surface is typical of tapped slag. So
the appearance of this piece has absolutely nothing related to
phosphorus content in bloomery iron itself. (I was unable to find the
source for this image.)
(at
9:52) “Between the 5th and
14th century, European sword smiths
created...”
Incorrect
dating :
“
Use of the process largely vanishes from western European blades
beyond the 11th century. ” (12)
The
term 'pattern welding' as is (correctly) applied here refers quite
specifically to the 'multiple twisted core' construction method. This
is primarily a technique found in Northern Europe, the finest
examples found in Anglo-Saxon burials (Sutton Hoo, c 625 CE). Many
simpler (two core) samples are known from the Norse / Viking Age,
including the well documented '+ULFBERTH+' inlayed swords.
(at
10:05) “Phosphoric iron from bog ore was used specifically
as one of the contrasting metals specifically because it corroded at
a different rate that the plain iron and steel layers.”
After
considerable attention was paid to the difference in potential
hardness, the assumption is made that it was for simple differences
in contrast appearance that phosphoric iron was incorporated into
these distinctive pattern welded blades. This ignores the reality
that combination of iron and 'steel', and likely phosphoric content
iron / steel, were combined primarily for their differing functional
qualities (as discussed above). It is clear that additionally
the complex patterns of alternating twisted and straight line
sections to the core bars was done intentionally for decorative
effect
(at
10:15)
Totally
incorrect image. There is a close up of a etched (modern?) knife
blade surface, which is clearly not 'multiple layer, twisted
core', but in fact wootz. (13)
Wootz is a material made not from welding differing layers
together, but is a crucible process, where a combination of
carbon content irons are melted together into a solid ingot,
where carbon varies as the mass cools. The method was certainly known
and concentrated to the Mid East, specifically in Damascus. The
evidence that these complex production methods were known and
undertaken in Europe is controversial at best.
(from
10:18) “..the phosphoric iron remained bright, while the
other metals darkened. ... The pattern welded technique relied on the
corrosion resistance of bog iron to make the patterns visible.”
Misleading.
What is hidden here is that all the metals involved were
smelted from bog iron ores. The ancient smiths had deliberately
chosen bars from blooms with differing characteristics, almost
certainly made in widely different locations (so trade items). In
fact, the primary difference in colouration will come from variations
in carbon content between the starting bars. Carbon containing
alloys will react to acid solutions, turning dark. A carbon and
phosphorus free wrought iron reacts inconsistently, microscopic
contained slag (so glass) layers becoming visible resulting in a
light grey 'ropy' texture. (14)
Much
is detailed in this section about the historic use of phosphorus rich
bog ores. There is no mention made of the significant difference in
understanding of the underlying chemistry between the Medieval
period (so here specifically Anglo-Saxon and Norse) and modern (post
Victorian) metallurgical science. It is clear that those historic
sword smiths understood the difference in working metal quality
between individual iron bars, and if proven trade patterns is any
indication, quality between iron created from ores in differing
locations. But they had no understanding of just what make bars from
one location different than those from another. The individual skill
and experience of the specific iron master would also be a major
functional difference between separately produced bars.
(starting
10:50)
As I am intimately familiar with L'Anse aux Meadows NHSC
(LAM) and the iron smelt there by the Norse circa 1000 CE, I have
considerable insight here. (15)
First
off, the use of AI narration showed another of it's problems here.
Three times the location name was given, each time pronounced not
only incorrectly, but also differently each time. (From the French?)
This segment is based on the original interpretations of the
archaeology, from the later 1960's. These impressions are quite
dated, and certainly not the current understanding of the site,
particularly related to the single iron smelting attempt made there.
The suggestion that there was a direct link from boat repairs to a
requirement to smelt iron has certainly been challenged in the light
of direct experimental results. (16)
(from
11:14) The building shown is a more fanciful interpretation,
actually located at nearby Norstead (which is best described as a
'historic based entertainment'), not the archaeologically more
accurate structures recreated at LAM.
(at
11:20) “Archaeologists have recovered...along with
15 kg of slag”
Incorrect.
The actual amount of iron slag recovered archaeologically related to
the iron smelt was in fact only 10 kg, the amount of 15 kg was at
best a semi-educated guess.
(at
11:30) There is a significant difference between a nail and a
ship rivet. (Noting nails and rivets are distinguished earlier) The
artifact nail described, has a square cross section and long taper,
and was found at one of the house entrances, suggesting it was part
of a wooden door. Rivets on the other hand, are round cross section,
straight shafts and are peened over on both ends.The rivets found at
one of the small workshop buildings were cut into two pieces,
indicating that they had been pulled out of what was most likely a
smaller fishing boat boat sized vessel. It is true that the single
nail described did have an elemental composition matching that of
roasted bog ore recovered at the smelting furnace (not the slag, as
is stated).
(from
11:36) “A single Norse longship required approximately 7000
nails, forged from around 400 kilograms of finished iron. Producing
that much iron from bog ore means a blacksmith would have to process
roughly thirty tons of raw ore.”
No
idea where those numbers come from? (It may be a single incorrect
value has distorted all the following ones?) Run the math here:
“...approximately
7000 nails...”
This
number vastly inflated. Both the archaeology and the physical
replica work at the Viking Ship Museum in Roskilde, Denmark (17)
indicates that the construction of a Norse knarr (ocean going
freight ship – like the ones travelling the North Atlantic) would
in fact require closer
to 1200 - 1500 rivets paired with roves). Of
course recovered Norse ship hulls range from small light fishing
boats through to massively long warships.
“...forged
from around 400 kilograms of finished iron.”
Take
400 kg / 7000 = 57 gm each. Importantly here, this is again not
counting the weight of all the matching roves. It would be
reasonable to assume 10% loss in the conversion of staring
'currency bars' into forged rivets (18),
which indicates 52 gm per rivet. The physical size of an individual
rivet varies considerably, shaft diameter and length dependant on
the type of timber / thickness of the planks. The largest and most
complete ship rivet recovered at LAM, #60, was 8 mm diameter, which
suggests a replacement rivet at about 50 gms (again not including
the required rove). (19)
With the more
realistic total of rivets at 1500, closer to 85 kg of bar iron
would be required for the rivets
(again noting the required roves are not included).
“...thirty
tons of raw ore”
Given
that all the other measurements used in this presentation are in
metric units, this is likely to mean metric tonnes = 1000 kg, so a
total of 30,000 kg?? This clearly
must be completely
wrong.
Experience
undertaken here suggests a reasonable return at the bloom to bar
phase is about 65%. This would suggest that the starting blooms
would weigh roughly 615 kg. (20)
So,
30,000 kg of ore used to produce 615 kg of blooms would mean the
smelting portion would have a yield at an unrealistic overall
yield of only 2 %
Using
a more realistic yield number (average from our modern experiments)
at 22% bloom from ore the blog
iron amount required for that 615 kg would be 2820 kg.(21)
(At roughly 3 metric tonnes, this links back to that incorrect ore
number?)
Taking
the more realistic total number of rivets, losses, and yields
together, 1500
finished rivets from 85 kg of bar iron requires approximately 425
kg of bog iron ore.
(Again noting that the elemental iron content of the starting ore
an important variable here – see below.)
(from
12:02) “ The
Settlement at L'Anse aux Meadows was positioned immediately adjacent
to a peat bog, it was not a co-incidence.”
There
are a large number of practical and strategic reasons why the the
Norse Greenlanders would have chosen grass covered marine terrace at
the foot of Epaves Bay. The
proximity to a peat bog is so low on the list as to be
inconsequential (most especially in this part of North Newfoundland,
which is widely covered with peat bogs).
(from
12:21) “Analysis
of the L'Anse aux Meadows slag shows that significantly more iron
could have been extracted from the ore than actually was, suggesting
the workers there were competent, but not expert smelters. This
supports an important point. Bog ore smelting was common knowledge,
not a specialized trade secret specialized to master smiths.”
“Analysis
of the L'Anse aux Meadows slag...”
Clarification:
From the original report, three slag samples were analyzed,
averaging with a combined remaining elemental iron content of 61.3
%.
It
is also worth noting that the bog ore sample analyzed from LAM
contains almost no elemental phosphorus at all.
“...significantly
more iron could have been extracted from the ore than actually
was...”
Earlier
in the video, the (false) opinion is given that early bloomeries
operated at 10 to 20 % bloom yield from ore weight (as discussed at
some length previously). Using
the most simplistic method of (estimated) bloom = output against
recovered slag + bloom = input , the yield at LAM would be 3 kg out
/ 18 kg in = 16.7 % yield – which is well within the return range
as stated. The two statements made contradict each other.
The
actual analysis of the roasted bog iron ore recovered in the
furnace hut at LAM has an elemental iron content of 69.8 % (close
to the theoretical maximum for pure Fe2O3). Comparing with the slag
content, this puts the amount of elemental iron 'missing' (so into
the created bloom) at only 8.5 % (22)
Against this pure ore, this is an extremely poor return on the
effort.
“Bog
ore smelting was common knowledge...”
This
assumption, common to Norse researchers 50 years ago, has most
certainly been dis-proven repeatedly by experimental archaeology
since those days. (23)
Compare with a (very simplified) example : 'everyone' in the
West knows 'how' to to make an apple pie. But you certainly would
not enjoy eating most of those! Bloomery iron smelting is
significantly more complex.
(from
12:40) “ From Russia to the Revolution “
In
this segment, centuries of technical innovation, industrial
production from the Renaissance to the early Victorian periods, and
individual locations ranging from Russia to Civil War United States
of America, are squashed together simplistically. Although it was
true that easy to access bog iron ores were often the first types
utilized in any given location (where even available), these small
deposits could not supply the Industrial Age furnaces. This is most
specifically true in Colonial era USA, despite the attempt in the
narrative to imply bog iron sources were the primary supply.
(at
13:05) “The first known iron mines in North America reportedly
in operation by 1578, at St Johns Newfoundland, were bog iron
operations.”
This
is incorrect in almost every detail. The earliest record of iron ore
observed on Bell Island is from 1578. It was not until 1628 the first
samples were sent back to England for assay. It would not be until
1895 that the Wabana mine was established. This was a hard rock mine,
not a surface deposit of bog iron ore. (24)
(from
13:10)
Distortions.
-
The Saugus complex primarily a blast furnace operation, producing
high carbon cast iron, with secondary production of bars using the
finery forge methods. The ore used was limonite, not a mud like
primary bog iron ore (as is described earlier in the presentation).
(25)
-
The Nassawango Iron Furnace did use local bog iron ore but again was
a blast furnace for production of cast iron, and was both 200 years
later and very short lived (the dates given are incorrect) :
-
(at 13:35) “...specifically
valued for it's corrosion resistance.”
From
the early 1700's through to about 1840, iron production in the USA
was centred on charcoal fired smelters through south eastern New
Jersey and southern Pennsylvania. This was due to a combination of
available primary bog ore and limonite deposits, forest cover for
charcoal supply, and terrain providing ample water power. (27)
A large number of both bloomeries for bar production and blast
furnaces for cast iron were in operation. I could find no specific
mention of specific 'corrosion resistance', and this sweeping
generality is highly unlikely due to the large number of furnaces
over such a wide spread area and so long a time period. If anything,
the primary complaint visible in descriptions of the times was the
lack of consistency in quality in wrought iron bars.
(
from 13:50)
“The
shift began with the start of coke fired blast furnaces in the 18th
century. These furnaces could process mined ore, particularly
hematite and magnetite at vastly higher temperatures and volumes than
bloomery furnaces could handle”
There
are several distortions.
-
Although not European, the first use of coke for iron smelting is
documented from China dating to as early as 1000 CE, where charcoal
fired cast iron furnaces are known from as early as 100 BCE. The
earliest patent for using coke for iron production (in England) dates
to the later 1500's. The statement as given appears to refer to
Abraham Darby (at Coalbrookdale, England) who established a coke
fired cast iron blast furnace in 1709 CE. (28)
-
The earliest cast iron blast furnaces in Europe date at least to
1200, possibly as early as 1100 CE. (29)
It was simpler to create plates of cast iron from ore, then use a
secondary process, the finery, to burn out the contained carbon down
to forgeable iron bars. The key here is the introduction water
powered, more massive, bellows over human operated sizes, allowing
for higher operating temperatures.
-
Japanese tatara furnaces, in use from at least the middle 500's CE,
where developed specifically to use magnetite in the form of iron
sand. (30) There are certainly regions where rock iron ores, in the form of ancient 'banded iron' were available as surface deposits, and used.
(from 14:20)
This
segment strings together a number of elements, that although may be
true individually, do not in fact support each other.
-
In truth, was is totally ignored here is the massive difference
between Early Medieval (pre 1000 CE) production rates and individual
access to iron (steel) in our modern Western world. Scandinavia
during the later Dark Ages is the main comparison used in this
presentation. During that time and place, the average (lifetime) iron
share per individual within a farming household is roughly 2 kg each.
In comparison for the USA, for one year alone (2023), the steel
share per individual was 266 kg (31).
Clearly there is a vast difference here, and any attempt to equate
these two realities is nonsensical.
-
The intention throughout is to equate Early Medieval production and
bloomery created wrought iron to modern industrial steels with all
it's alloys and applications. The huge difference between corrosion
resistance, physical strength and potential hardness of modern
nickle content stainless steels and historic bloomery iron alone
proves the comparison being made is ridiculous.
(from
14:45)
“ The
United States alone experiences roughly 240,000 water main breaks
every year, the majority caused by corrosion.
“
-
The majority of these water main pipes will be made of brittle cast
iron, not modern steel as is implied.
-
This is important one reason for breaks is not corrosion,
but bursting in winter freezing and impact with machinery, which cast iron is susceptible to.
-
The major reason for breakage is simple aging of the long ignored
below ground infrastructure (in many cases over a century old).
(from
16:02)
As
might be expected, being directly involved with experimental
archaeology into specifically Norse / Viking Age bloomery iron
smelting, now for over 25 years, again I consider I have insight
here. (32)
(from
16:22)
The
furnace shown in the background during the segment about Hurstwic,
although truthfully that used for their two smelt demonstration
program on Iceland, actually does not
conform what is known from archaeology about Icelandic turf built
furnaces. The furnaces recovered at Hals by K. P. Smith show a
conical
stack construction method. (33)
(from
16:30)
“Successfully
producing high quality iron...”
Although
this claim has been made repeatedly in promotions by Hurstwic, there has never been any description of what 'high quality' means here. There
has only been minimal information made available on the Hurstwic web
site, indicating extremely low smelt yields, with no mention of smelt
sequence, bloom densities, carbon contents, or bloom to bar losses
that would support this assertion.(34)
(at
16:50)“
...and
in some measurable respects, superior to modern materials for
specific applications, were corrosion resistance is more important
than volume.”
This
bold statement is made, with absolutely no
evidence what so ever
to support it.
Clearly
modern alloy steels, with all the specifically engineered
characteristics possible eclipse bloomery iron. (35) There may be some truth to the endurance of later medieval ornamental wrought iron grills and gates. But the comparison here is being made to simple modern mild steel, not available intentionally corrosion resistant alloys. There is no consideration being made to the problems of weight over strength or forging over machining.
(from
18:00)
“The
EXARC Journal, an international experimental archaeology publication,
contains step by step reconstructions of Medieval bloomery processes.
The Hurstwic group's research on Icelandic smelting is publicly
available.”
Again,
personal bias. One of the publicly available complete guides to
operating a historic bloomery furnace on the EXARC Journal web site
is in fact my own (as well as a number of other related articles
there). (6B)
In truth, the information on Hurstwic's web site does not present a
working method that could be followed, and they have never
contributed to EXARC (or the larger Early Iron community).
(from
18:35)
“...in
Newfoundland... his ship needed nails... produced the metal he
needed.”
This
is the second time this mis-conception is repeated (and was critiqued
earlier) :
Lief
Eriksson (if the Sagas are accurate) was a second generation,
professional expedition leader. It is inconceivable that the
Greenlanders would undertake the extremely long voyage into the
western unknown (a sailing distance equal to travel all the way back
to Norway!) without carrying enough ship rivets for emergency
repairs.
The
weight of specialized iron tools required to smelt iron exceeds the
3 kg weight of damaged rivets that were found. Surely if needed, the
large iron tongs necessary for bloom handling would be converted
into rivets? Although a basic set of blacksmithing tools might have
been standard ship repair equipment, the different tools for iron
smelting would not be.
Although
there is bog ore in Greenland, there is no timber suitable for
making the large amounts of charcoal required for a single smelt.
Iron objects and working bars would be the primary import items
during the 400 + year life of the Greenland colony. The ability to
produce iron in Vinland would be yet another 'boast' of the richness
of the newly explored region. (16B)
(from
18:55)
The
nail illustrated is in fact clearly a modern, cylindrical shaft,
'wire' nail, made of mild steel. Not one hand forged, or made of
bloomery iron, or a Viking Age ship rivet - much less one of the
artifacts recovered at LAM. (36)
(from
19:25)
Again,
the video presents it's core thesis, that early Medieval bloomery
iron was superior to modern steels. Again the figure of world wide
losses due too steel corrosion is raised.
As
detailed earlier, the massive difference in personal share of metal
that is certainly two, if not three, magnitudes greater in our
modern world, is not considered.
There
is no comparison of the yearly production value of new steel world
wide, again massively larger than the corrosion lost number given.
The
is no accounting for the overall production costs, in terms of
labour and expended materials, in small scale bloomery iron making. (37)
This
is such a stunning distortion of the reality, it stretches beyond
disingenuous to the intentionally completely false.
Part
Three : 'Trouble with AI'
Although
the location source for Medieval Wisdom is listed as USA, the
repeated stress on research and experiments from Hungary suggests the
true source for these productions is in fact Eastern Europe. The use
of computer synthesized narrations for English, obviously AI
generated from a written text, also suggests a disconnect here as
well. This does raise questions as to the true purpose of these
videos, especially considering the extremely rapid production rate.
Potential data mining (especially directed to those who have
subscribed) comes to mind.
As
I checked individual statements, it was clear that although the
narration text did not directly copy available documents, the
content certainly directly paraphrased, particularly from Wikipedia
entries. Commonly source information was used in part or out of
context, particularly when doing so would strengthen the underlying thesis being presented.
The
overall form of the final text has the same kind of feel as summaries
produced by the currently available AI language systems like
Grammarly or ChatGPT. These systems have no
understanding of the content they base their output from, they
merely string together pieces into something that superficially
appears to be an average description. Given the huge number of
distortions and mis-conceptions presented in this video, it is almost
certain that the narrated text itself is completely AI generated.
It
was worth noting as checked base facts I would often see the
individual images used as the background video. Often still images
had been converted into a form of 'drift over' panning to impart
motion where none originally existed.
In
preparing this critique I have spent a rough total of 36 hours (12
day sessions of typically 3 hours each). This could be seen as a
reflection of my own (declining) concentration and working speed? I
am however quite familiar with the topic area. Over that same number
of days (12) 'Medieval Wisdom' has produced a further 12 similar
video presentations.
I
see 'Medieval Wisdom' is a clear reflection of an ongoing
cultural shift related to the internet and content ownership, and the
concept of knowledge itself.
Those
who initially stocked the world wide web with information content (so
1990's) expected to retain ownership of that content (although
perhaps ill considered?). Into the 2010's a new generation (who
always had an internet) began to take the opinion that posting
anything to the web resulted in total loss of ownership and control
(despite clear statements of copyright, watermarking). Over the
last year especially, the (alarming) growth of AI internet mining
systems and content generation softwares is making outright theft
almost standard. Certainly the entire concept of 'originality' and
'authorship' is being discarded.
I
have become increasingly concerned with two currently cultural
factors that magnify each other :
Lack
of critical evaluation skills, most especially among younger
people.
Explosion
of slickly manufactured, but factually incorrect, presentations.
As
someone deeply involved in traditional skills (to the point of
recovering lost historic methods), I see a constant eroding of
personal experience as a learning method.
(And
yes, Gentle Reader, I am fully aware I am 'pissin' in the wind'.)
Notes :
Any
internet links given below were current at the time of writing
(February 2026)
The
use of Wikipedia as a reference source is intentional, illustrating
how little depth would be required to check specific points.
See
: Wikipedia - Bloomery : https://en.wikipedia.org/wiki/Bloomery
Historical
Marker Data Base : https://www.hmdb.org/m.asp?m=3922
The
entry in Britannica referenced here is only 4 lines long, and
certainly makes no attempt to define what is meant by 'Economically
useful'.
See
: Britannica - Bog Iron Ore :
https://www.britannica.com/science/bog-iron-ore
See
: Ekomuseum
Bergslagen :
https://ekomuseum.se/en/vill-du-veta-mer/jarn-ur-rodjord-och-sjomalm/
Image
source : Medieval Histories :
https://www.medieval.eu/smelting-iron-viking-way/
There
are a number of basic method guides available :
Sauder
& Williams, 2002, “A Practical
Treatise on the Smelting and Smithing of Bloomery Iron”,
the Historical Metallugical Society (PDF
here)
Markewitz,
2012, “ 'But If You Don't Get Any
IRON...' Towards an Effective Method for Small Iron Smelting
Furnaces”,
the EXARC Journal (PDF
here)
Over
100 personal iron smelts, the total of previously graded for size
charcoal expended over a typical 25 kg ore weight smelt averages 45
kg. (see note 9)
Refer
to the extensive documentation : Experimental Iron Smelting :
https://www.warehamforge.ca/ironsmelting/index.html
See : Caherconnell Bloom Forging :
https://warehamforgeblog.blogspot.com/2023/12/caherconnell-bloom-forging-1.html
See : Markewitz & Peterson, 2025, “ 'How Dense are You?' -
Recording Bloom Density from experimental iron smelting”,
Kindle / Amazon
See Markewitz & Peterson, 2025, “ '20 Years Before the
Blast' - Experimental Bloomery Iron Smelting 2001 – 2024”
Kindle / Amazon
For a good overview of the topic of differing forms of both layered
and crucible historic blade materials see : Lee A. Jones, ND,
“Blade Patterns Intrinsic to Steel Edged Weapons”, web site :https://www.vikingsword.com/patterns.pdf
See a description of 6 bloomery iron knives : Bloom Bar to Blade :
https://warehamforgeblog.blogspot.com/2023/06/bloom-to-bar-to-blade.html
See : Wikipedia - Wootz Steel : https://en.wikipedia.org/wiki/Wootz_steel
The primary reference used here is : Ingstad, 1975, “ The
Discovery of a Norse Settlement in America”,
Universitetsforlaget
The original assessment (boat rivets needed = smelt iron) was made
by researchers (understandably) not familiar with the process of
bloomery iron smelting in the early 1960's. Iron production was
estimated by comparing bog ore iron concentration with that
remaining in the recovered slag, multiplied by a rough guess of the
total slag created (only a fraction had actually been recovered).
This very roughly estimated number, 3kg of possible iron bloom, was
compared to the mass of broken boat rivet fragments recovered, the
same weight. Worth noting here that this comparison was not
determined by adding up how many whole rivets would be required.
Significantly no allowances were made for the considerable losses at
the bloom to bar phase, and additional losses at bar to rivet
forging. See : Markewitz, 2025, 'Iron Smelting in Vinland :
Converting archaeological evidence to a practical method',
Kindle / Amazon
See : Ottar (Skuldelev
1) :
https://www.vikingeskibsmuseet.dk/en/professions/the-boat-collection/ottar
This based on my own 40 years experience as an artisan blacksmith,
including considerable work with self made bloomery iron.
As a check here, The weight of a piece of modern mild steel long
estimated enough to form a replacement for the LAM rivet (7 mm dia.
By 8 cm long) weighs 42 gm – before any forging loss.
Having amassed a considerable number of blooms over the last 25
years, Neil Peterson and myself are proceeding to render these down
into working bars. As this project has continued, we are making
increasingly detailed records of number and types of forge welding
steps, along with the number of major folds and resulting carbon
content variations. The plan is to format this data into a
publication within the next 12 – 18 months.
Looking at the totals for our own work over the last 25 years :
2,590 kg of (mixed) ore yielded 485 kg of raw blooms. This during
the process of learning the iron smelting process from nothing, and
specifically testing many variations on build, technique, and air
systems. See Markewitz & Peteson, 2025, “ '20
Years Before the Blast' – Experimental Bloomery Iron Smelting 2001
– 2024”, Kindle / Amazon
See : Rosenqvist, 1974, “Material
Investigations”, in Ingstad,
1974, “The Discovery of a Norse
Settlement in America”,
Universsitetsforlaget
I would refer you to the published work of Peter Crew, Thijs van de
Mannakker, Lee Sauder, Jens Oleson, Micheal Nissen, or myself, among
others.
See
: References :
https://www.warehamforge.ca/ironsmelting/smeltlinks2.html)
See : Iron Mines of Bell Island :
https://www.heritage.nf.ca/articles/economy/bell-island-mines.php
See : Saugus Iron Works :
https://www.discoveramericablog.com/post/saugus-iron-works
Furnace Town Historic Site, ND, “Nassawqango
Iron Furnace”, web site :
https://www.furnacetown.org/iron-furnace-1
See : Philadelphia Encyclopedia - Iron Production :
https://philadelphiaencyclopedia.org/essays/iron-production/
See : Wikipedia - Coke : https://en.wikipedia.org/wiki/Coke_(fuel)
See : Wikipedia – Blast Furnace, History :
https://en.wikipedia.org/wiki/Blast_furnace#History
See : Wikipedia – Tatara :
https://en.wikipedia.org/wiki/Tatara_(furnace)
World Steel Association, 2025, “World
Steel in Figures”, web site :
https://worldsteel.org/data/world-steel-in-figures/world-steel-in-figures-2024/
I freely state my bias when it comes
to the work done by Hurstwic. I in
fact gave members of this group their initial instruction into Norse
bloomery iron smelting in June 2018, at which point my own team had alreadyundertaken eight smelts in our own Icelandic / Hals experimental
series (starting in 2007). I had been part of the same Hurstwic /
Icelandic project illustrated in 2019, only to be withdrawn from
this (with only two weeks remaining before flight date).
See : Smith, Kevin P., 2005 "
Ore, fire, hammer, sickle: iron production in Viking Age and Early
Medieval Iceland", n De
Re Metallica: Studies in Medieval Metals, AVISTA Studies in the
History of Medieval Technology, Science, and Art, Volume 4,
edited by Robert Bork et al, Ashgate Press
Yields for the three smelts undertaken listed at 4.7 / 10.3 / 2.0 %.
See
: Hurstwic – Iron Smelting in Viking Age Iceland :
https://www.hurstwic.com/research/iron/index.html
As a working artisan blacksmith with decades of experience, the one
advantage wrought iron has over modern steels is not corrosion
resistance, but in wrought iron's ease of forging, particularly for
aggressively shaped art works.
See: Lessons from the Viking Age – ship rivet :
https://www.warehamforge.ca/ENCAMPMENT/IMAGES/lessons/rivet.jpg
For the creation of the often described 3 kg bloom estimate at LAM,
our own team expends :
30
kg dry powdered clay, 30 kg sand (gathered locally?)
50
kg rough charcoal (timber cut, split, converted / not including
losses)
20
kg ore – depending on quality (gathered, roasted / not including
losses)
In the modern experimental smelting here, the clay, charcoal and iron oxide
for ore analog is all purchased, at a rough cost of $300 CDN (2025).
The
labour required for building a single furnace, preparing all the
materials for smelt, operating the smelt through to a single
compaction cycle averages averages about 52 hours. If Norse style
bellows are providing air, add an additional 32 hours. At current
Canadian minimum
wage (2026, Ontario) of $17.60 / hour the labour cost (for full Norse method)
comes to a rough cost of $1480 CDN.
This
places the modern production cost of
a single kg of raw bloom at $590
CDN (300 + 1480 / 3). This is not considering costs related to
specialized skilled labour, and additionally time, consumed
materials and physical losses at a the bloom to bar phase. Compare
this with the 2025 market price for modern industrial hot rolled mild
steel of 0.70 $ CDN per kilo.
Posts
