27-08-2026, 07:02 PM
Hello,
I wanted to raise some potential issues that I've had with the carbon dating of the VMS. Before I start, I want to make abundantly clear that I accept the dating for the individual leaves of parchment as reported, and that this is NOT an "anti carbon dating" post in any way.
I believe that some of the core assumptions used to come to the final dating range do not necessarily hold up to scrutiny. Before I do so, I want to make clear the following things that I have no issue with:
- I see no issue in the handling or testing process for any of the individual samples
- I do not disagee with any of the individual dating results for any of the samples
- I believe that the results were correctly calculated after the core assumptions were applied (I ran the numbers myself as well and got pretty much the same range)
---------------------------------------------------------------------------------------------------------------------------------------
With that out of the way, let me explain what I mean. My issues stem from the combination of the invidual dating results into a smaller range using a statistical process. As I mentioned above, I believe that this statistical process was done entirely correctly after the core assumptions were applied. But, what were those assumptions that justified doing so?
1) The individual leaves of parchment are part of a singular "whole" object, and thus can all be assumed to be samples from that same "object"
What is the justification for that to be assumed as correct? As far as I can tell, the main reason to accept this is that the entire VMS contains "voynichese". That's pretty much it. As i've frequently come across while looking through manuscripts, it is common for there to be a seemingly random inserted page, sometimes far older, sometimes far younger within the binding. And often the only discernable way to know that it is dated differently is the content: perhaps the handwriting is different, or the language is different. We have no such luxury with the voynich manuscript, as we cannot understand the content.
Okay, but isn't this part of the reason why samples were taken from different sections? Yes, but there are FAR too few samples to conclusively say that there are no outlier leaves. If the manuscript was indeed previously loose sheets in a different order, it's completely possible that there are some older or younger leaves scattered within the sections. For instance, say that someone began writing the VMS in 1380AD, and completed around 4 pages, but never finished the rest or died. Then, a student or family member continued the work in the same style, with the same writing rules, later in 1420AD. When the book was rebound later, those 4 pages would not have stood out from the rest, and would now be inserted at random within a section.
So seperately, it is possible that if we were to take 4 more samples from 4 other leaves, we may happen to get completely different results. That is the nature of seperate pieces of parchment, and more importantly that is the nature of those sheets NOT being a single object. Let me present this another way: if 20% of ALL folios were truly from 1350AD, and 80% were from 1420AD, the chance to pick 4 samples from 1420 at random is around 40%.
2) The VMS was written and produced in a short period of time
I don't understand the evidence for this. How can we possibly know?
As I mentioned above, there are countless instances of composite manuscripts containing works from different time periods. Quite often, there is no real discernable difference between the quality of the parchment - it is only the content that changes. Sometimes, I come across examples where the text was written at a certain time, and the illumination didn't take place until decades later. Take this example, where the text was written in 1425-1450AD (or maybe slightly earlier), but the illuminations date later, with some dating to around the last quarter of the 15th century: You are not allowed to view links. Register or Login to view.
If you were to take 4 samples from that manuscript, how many would come back saying 1490AD?
The biggest issue with this assumption is that it effectively stifles research into anything post 1450AD, when it is entirely possible that sections of the VMS were written after 1450AD. We can point at certain aspects that potentially date the VMS, such as the clothing, or the zodiac signs, or the swallowtail merlons. Okay, but that dates those pages, not the entire VMS. It actually doesn't date those pages with certainty either; there are plenty of examples of almost 1-1 copies of illustrations over decades.
Seperately, this assumption itself implies something: that the works in the VMS must date to the same time period as the parchment. This, I believe for obvious reasons, is not a reasonable standpoint. What is to say the VMS author was not sourcing from an italian manuscript from 1320AD, and seperately sourcing another french manuscript from 1410, and then also sourcing a german manuscript from 1250AD? Sourcing works from 1250AD sounds unlikely, but to do so in 1430AD would be almost exactly like us sourcing from Charles Dicken's Great Expectations, from 1860. Not farfetched in the slightest.
------------------------------------------------------------------------------
The 14th century f68 peak
A seperate issue that comes forth after doing this statistical process is the dismissal of the huge early-mid 14th century peak in the f68 result. This is effectively brushed aside by aggregating the results of the different pieces of parchment. Is this proper? It seems entirely improper to state that a single piece of parchment has a 42% chance of being dated to 1308-1362, but then to take 3 other seperate samples, state that they are the same object as f68 (when they are not), and then say there's a 95% chance that the whole thing dates to 1404-1435.
This is basically saying "yeah, theres that huge peak there, but these other pages dont have the peak, so its probably worth ignoring". Why can't one page genuinely date to 1330, and another genuinely date to 1420?
Again, if we had chosen 4 other pieces of parchment, and each had the same ~50-50 peak distribution as in f68, there would have to be a huge discussion on which peak was more likely. But because we only found one, we dismiss it entirely. It's not right.
If the reality is genuinely that some pages date to ~1330, and others date to ~1420 (coincidentally at the same levels of atmospheric carbon), we would expect to see peaks in both for both sets of pages.
--------------------------------------------------------------------------------
Conclusion
Basically, this is what I am making the thread to state:
- The dating range of 1404-1435AD is based on what I believe are potentially incorrect assumptions.
- Based on the results, we should probably have a likely working date-range of ~1385-1458AD
- More carbon testing could potentially find more pages similar to f68, with a large peak in the mid 14th century
- To say that the VMS was written in a short period of time, or saying that the VMS is dated fom 1404-1435, even if true, puts unneccesary restrictions on various aspects of research.
- We should be treating works from the 9th-15th century as possible direct inspirations for the content in the VMS, instead of broadly dismissing most as probably not relevant.
- We should not presume that the content cannot have been post-1435AD
I know that making this thread will probably not win me any favours, and i've been sitting on it for a while now debating whether to post. But if anything comes across as an attack on other researchers and their work, please do not take it that way. I don't mean for that to be the case at all, and I have huge respect to those that have done that work.
I wanted to raise some potential issues that I've had with the carbon dating of the VMS. Before I start, I want to make abundantly clear that I accept the dating for the individual leaves of parchment as reported, and that this is NOT an "anti carbon dating" post in any way.
I believe that some of the core assumptions used to come to the final dating range do not necessarily hold up to scrutiny. Before I do so, I want to make clear the following things that I have no issue with:
- I see no issue in the handling or testing process for any of the individual samples
- I do not disagee with any of the individual dating results for any of the samples
- I believe that the results were correctly calculated after the core assumptions were applied (I ran the numbers myself as well and got pretty much the same range)
---------------------------------------------------------------------------------------------------------------------------------------
With that out of the way, let me explain what I mean. My issues stem from the combination of the invidual dating results into a smaller range using a statistical process. As I mentioned above, I believe that this statistical process was done entirely correctly after the core assumptions were applied. But, what were those assumptions that justified doing so?
1) The individual leaves of parchment are part of a singular "whole" object, and thus can all be assumed to be samples from that same "object"
What is the justification for that to be assumed as correct? As far as I can tell, the main reason to accept this is that the entire VMS contains "voynichese". That's pretty much it. As i've frequently come across while looking through manuscripts, it is common for there to be a seemingly random inserted page, sometimes far older, sometimes far younger within the binding. And often the only discernable way to know that it is dated differently is the content: perhaps the handwriting is different, or the language is different. We have no such luxury with the voynich manuscript, as we cannot understand the content.
Okay, but isn't this part of the reason why samples were taken from different sections? Yes, but there are FAR too few samples to conclusively say that there are no outlier leaves. If the manuscript was indeed previously loose sheets in a different order, it's completely possible that there are some older or younger leaves scattered within the sections. For instance, say that someone began writing the VMS in 1380AD, and completed around 4 pages, but never finished the rest or died. Then, a student or family member continued the work in the same style, with the same writing rules, later in 1420AD. When the book was rebound later, those 4 pages would not have stood out from the rest, and would now be inserted at random within a section.
So seperately, it is possible that if we were to take 4 more samples from 4 other leaves, we may happen to get completely different results. That is the nature of seperate pieces of parchment, and more importantly that is the nature of those sheets NOT being a single object. Let me present this another way: if 20% of ALL folios were truly from 1350AD, and 80% were from 1420AD, the chance to pick 4 samples from 1420 at random is around 40%.
2) The VMS was written and produced in a short period of time
I don't understand the evidence for this. How can we possibly know?
As I mentioned above, there are countless instances of composite manuscripts containing works from different time periods. Quite often, there is no real discernable difference between the quality of the parchment - it is only the content that changes. Sometimes, I come across examples where the text was written at a certain time, and the illumination didn't take place until decades later. Take this example, where the text was written in 1425-1450AD (or maybe slightly earlier), but the illuminations date later, with some dating to around the last quarter of the 15th century: You are not allowed to view links. Register or Login to view.
If you were to take 4 samples from that manuscript, how many would come back saying 1490AD?
The biggest issue with this assumption is that it effectively stifles research into anything post 1450AD, when it is entirely possible that sections of the VMS were written after 1450AD. We can point at certain aspects that potentially date the VMS, such as the clothing, or the zodiac signs, or the swallowtail merlons. Okay, but that dates those pages, not the entire VMS. It actually doesn't date those pages with certainty either; there are plenty of examples of almost 1-1 copies of illustrations over decades.
Seperately, this assumption itself implies something: that the works in the VMS must date to the same time period as the parchment. This, I believe for obvious reasons, is not a reasonable standpoint. What is to say the VMS author was not sourcing from an italian manuscript from 1320AD, and seperately sourcing another french manuscript from 1410, and then also sourcing a german manuscript from 1250AD? Sourcing works from 1250AD sounds unlikely, but to do so in 1430AD would be almost exactly like us sourcing from Charles Dicken's Great Expectations, from 1860. Not farfetched in the slightest.
------------------------------------------------------------------------------
The 14th century f68 peak
A seperate issue that comes forth after doing this statistical process is the dismissal of the huge early-mid 14th century peak in the f68 result. This is effectively brushed aside by aggregating the results of the different pieces of parchment. Is this proper? It seems entirely improper to state that a single piece of parchment has a 42% chance of being dated to 1308-1362, but then to take 3 other seperate samples, state that they are the same object as f68 (when they are not), and then say there's a 95% chance that the whole thing dates to 1404-1435.
This is basically saying "yeah, theres that huge peak there, but these other pages dont have the peak, so its probably worth ignoring". Why can't one page genuinely date to 1330, and another genuinely date to 1420?
Again, if we had chosen 4 other pieces of parchment, and each had the same ~50-50 peak distribution as in f68, there would have to be a huge discussion on which peak was more likely. But because we only found one, we dismiss it entirely. It's not right.
If the reality is genuinely that some pages date to ~1330, and others date to ~1420 (coincidentally at the same levels of atmospheric carbon), we would expect to see peaks in both for both sets of pages.
--------------------------------------------------------------------------------
Conclusion
Basically, this is what I am making the thread to state:
- The dating range of 1404-1435AD is based on what I believe are potentially incorrect assumptions.
- Based on the results, we should probably have a likely working date-range of ~1385-1458AD
- More carbon testing could potentially find more pages similar to f68, with a large peak in the mid 14th century
- To say that the VMS was written in a short period of time, or saying that the VMS is dated fom 1404-1435, even if true, puts unneccesary restrictions on various aspects of research.
- We should be treating works from the 9th-15th century as possible direct inspirations for the content in the VMS, instead of broadly dismissing most as probably not relevant.
- We should not presume that the content cannot have been post-1435AD
I know that making this thread will probably not win me any favours, and i've been sitting on it for a while now debating whether to post. But if anything comes across as an attack on other researchers and their work, please do not take it that way. I don't mean for that to be the case at all, and I have huge respect to those that have done that work.