Jump to content
Seriously No Politics ×

This is where the Book of Mormon happened, according to AI


Recommended Posts

Posted
17 hours ago, Millcreek said:

What would be sufficient geographical information to decide where a fictional text occurred? 

How do I know you are not a fictional person?

Posted (edited)
13 hours ago, carbon dioxide said:

AI only knows what information is put in but does not really do well determining the quality of the information.  Garbage in you get garbage out.  The Book of Mormon is an English translation from non English speaking and writing people several thousand years ago and how exact is the carry over from one to the other?  Does AI take all of those issues into account?  I highly doubt it.

If you remember the story of Vincenzo Di Francesca, the Italian man who discovered a coverless Book of Mormon in a trash barrel on a street in New York City. The book had no introduction and no descriptions of its origins. Since Vincenzo had no clue where the Book of Mormon events took place, he wasn't bothered by the so-called anachronisms in the book. Anachronisms in the book only appear when we force a setting. For example, domesticated elephants are not anachronistic in 2500 BC, but they are anachronistic to Mexico in 2500 BC. Judean migrants fleeing Oman by boat are not anachronistic in 600 BC (see recent ancient DNA papers on Meditarranean DNA in Pattanam, India) but Judeans are seemingly anachronistic to 600 BC New York. The Book of Mormon mentions silk. Silk and domesticated silkworms are not anachronistic to the Book of Mormon time period, but they are anachronistic to the Americas during the Book of Mormon time period. Steel swords are not anachronistic to Nephi's timeline, but they are anachronistic to Nephi's timeline if we demand Nephi was forging his swords in Guatemala in the 6th century BC.

I'd like to know if there truly are anachronisms in the text, because to me it seems we create these anachronisms by forcing the text to have happened in a specific location. The text itself never claims a specific location. The methodology is to describe the contents of the book to AI, without all the external commentary and assumptions about its origins. If there happens to be convergence between the text and the real world, anywhere, a model can be constructed. 

Edited by Zosimus
Posted
4 minutes ago, Kevin Christensen said:

Seem a bit dubious to ask for a geographical issue by providing zoological information, and that without considering translation issues and temporal issues.  Computers are notorious for the Garbage in/Garbage out problem.

For comparison, Larry Poulson located all of the passages describing the Sidon and did a computer-based search of a 3d map of the Western Hemisphere for candidates that matched the description.  He found one.  It also happened to be southward of a location that used cement at the right time, and had two distinct civilizations rising and falling at the right times.  Later a geologist produced a book testing candidate locations for geographical information.  

Sorenson, of course, famously started by locating all of the geographical details, over five hundred passages, and creating an internal map.

What should be easier to find? Animal bones or a volcanic eruption.  What if you have the bones but not the volcano and not the civilizations?  All paradigm choice involves deciding which problems are more significant to have solved?  Paradigms should be testable and provided solutions not just some isolated pet issue but with breadth and depth.  John Clark observed that the overall trend for the Mesoamerican setting is towards resolution of problems.  More recently the LiDar surveys that had the potential to undercut Book of Mormon descriptions revolutionized Mesoamercan archeology overnight actually moved the overall picture toward the Book.

FWIW,

Kevin Christensen 

Canonsburg, PA

I've seen no resolution of the problems that arise with the Mesoamerican model. First of all, you have to turn the entire peninsula 90 degrees to get north to align. I know, there's some paper somewhere written about how west is actually north but it is not convincing. Even the Sorenson and Clark internal maps only resemble the Isthmus of Tehuantepec if you turn the entire geography 90 degrees. You also have to make the assumptions that elephants or mammoths were present in Guatemala in 2500 BC, and that the fine silk described in the book was not from silkworms but rather pineapple fiber. LIDAR results only complicate the Mesoamerican model IMO. it just shows that the Maya were far more numerous than we previously knew, making it less likely that Lehite scribes wouldn't have ever mentioned them and would have been able to maintain their reformed Egyptian script for 1000 years, leaving no trace of it on the walls of those newly discovered Mayan temples and highways.

Anyway, I'm not here to debate Mesoamerican modelers. That model has its purpose. I'd like to know if there's a model that answers the questions the Mesoamerican model and all existing models have not.

Posted
39 minutes ago, Kevin Christensen said:

Back in 2008 both Larry Poulson and Brant Gardner pointed out that Mesoamercans viewed North as a quadrant rather than as a vector, as a slightly flattened x rather than as as a  + as we do.  The criticism of being tilted is popular for rhetorical advantage, but is both dated and not accurate. 

I can assure you I didn't say its 90 degrees off so I could twist things for some rhetorical advantage. I honestly don't see it. If you do, I'd be more than grateful if you could explain it to me since I've found nobody else that could. Here's the map from Gardner's paper:

JrbMpuR.png

From the persepctive of a Nephite at the center of the X, how does the Gulf of Mexico become their East Sea? How does everything in the left quadrant become their Land Northward? I must be missing something critical to the argument and its not for lack of trying. I'd honestly appreciate someone explaining.

Posted

@Kevin Christensen mentioned John Clark's work on an internal map, so let's ask ChatGPT for some insight.

Zosimus: Attached is a rough sketch of a map of the land described in the book. 

The sketch is based on some geographical features mentioned in the book along with 7 transects that roughly estimate the shape of the land. The land is an hour glass shape, like an isthmus or peninsula, with seas on the west, south and east sides. The entire area should be roughly equivalent to the size of Israel. The most distinct geographical feature described in the book is a narrow neck of land that served as a defensive line during battles. The text says that it would take about 1.5 days for a soldier to walk across the narrow neck. 

The land south of the narrow neck should be large enough to sustain a civilization between 600 BC and 420 AD.

Given these dimensions, check each region for a landmass that could fit. Give a score to each region.

Y2MUsNP.png

GPT-4: Based on the map and description you've provided, we're looking for a region with the following characteristics:

  • An hourglass-shaped landmass or isthmus.
  • Surrounded by seas to the west, south, and east.
  • Roughly the size of Israel.
  • Contains a distinct, narrow neck of land that could be crossed by a soldier in about 1.5 days.
  • The southern landmass must be large enough to sustain a civilization from 600 BC to 420 AD.
  • Considering these criteria, we will score each region. Here are the scores:

Europe: 2/10

While Europe has peninsulas and isthmuses, none that match the size and specific shape described.

South Asia: 8/10

The Indian subcontinent has a larger size but includes narrow coastal plains that could match the description partially.

Southeast Asia: 7/10

This region has many narrow landmasses and isthmuses, like the Kra Isthmus, but they are generally larger than the size of Israel.

East Asia: 2/10

East Asia's coastal regions do not typically match the hourglass shape described.

Central Asia: 1/10

This landlocked region does not have the seas as described.

West Asia (Middle East): 9/10

This region includes the Sinai Peninsula, which could potentially fit the description quite well.

North Africa: 3/10

North Africa's coastal regions generally do not fit the described shape or size.

Sub-Saharan Africa: 1/10

The geography does not align with the described features.

Australasia/Oceania: 1/10

The geography of these islands and landmasses does not match the description.

North America: 4/10

There are isthmuses, like the Isthmus of Panama, but they do not match the size or shape described.

Central America: 4/10
Similar to North America, with regions like the Isthmus of Tehuantepec, but not a close fit.

South America: 2/10
Coastal shapes do not match the specific description provided.

The scores reflect the likelihood that each region contains a landmass fitting the detailed description. West Asia (the Middle East) scores highest due to its proximity to the size and geographical features described, including several narrow land connections between bodies of water that have historically served as significant transit and trade routes.

Posted

OK, there's not much point in continuing to include West Asia or Central Asia. Maybe its time to narrow down the field to the existing Book of Mormon geography models reviewed in the Youtube video. I'm happy to add any others that you all might have in mind. 

Zosimus: I have narrowed the field down to six unique geographies. These are as follows: Great Lakes in North America, Isthmus of Tehuantepec in Central America, Baja Peninsula in North America, Isthmus of Darien in Central America, Adam's Bridge in South Asia, Isthmus of Kra in Southeast Asia

Using the sketch of the land uploaded above, rank each of the six geographies based on how closely they match the sketch and description. The ranking will be from 1 (least likely to match) to 6 (most likely to match)

GPT-4: 

To rank the six proposed geographies based on their similarity to the sketch and description you've provided, we'll consider the following characteristics: the hourglass shape, the presence of a narrow neck of land, surrounded by seas, and a size comparison to Israel. Here's how they might rank:

Great Lakes in North America
The Great Lakes region does not have a distinct hourglass shape or a narrow neck of land suitable for military defense.
Rank: 1/6

Baja Peninsula in North America
The Baja Peninsula is a long, narrow strip of land rather than an hourglass shape and lacks a distinct narrow neck.
Rank: 2/6

Isthmus of Tehuantepec in Central America
This isthmus does have a narrow portion but lacks the hourglass shape. It's also significantly wider than the land area described.
Rank: 3/6

Isthmus of Darien in Central America
While it is an isthmus, its geography does not closely resemble an hourglass shape, and it is covered by dense jungle, making it less likely to have been a significant defensive line.
Rank: 4/6

Adam's Bridge in South Asia
Adam's Bridge is a chain of limestone shoals between Pamban Island and Sri Lanka. It does not match the hourglass shape or the size of the land described.
Rank: 5/6

Isthmus of Kra in Southeast Asia
The Isthmus of Kra is closer to an hourglass shape and has a narrow neck of land, potentially fitting the description of a defensive line. However, its size is larger than that of Israel.
Rank: 6/6

Based on the criteria, the Isthmus of Kra in Southeast Asia seems to be the most likely match as it has a narrow neck that could serve as a defensive line and somewhat resembles an hourglass shape in its wider regional context. However, none of the regions listed perfectly match the description, as the specificity of the hourglass shape and the size equivalent to Israel are unique and may not be found in a real-world geography. The ranking is therefore relative and based on available geographical information.

Posted
7 hours ago, Zosimus said:

with seas on the west, south and east sides

You forgot the north sea - Helaman 3:8

Quote

8 And it came to pass that they did multiply and spread, and did go forth from the land southward to the land northward, and did spread insomuch that they began to cover the face of the whole earth, from the sea south to the sea north, from the asea west to the sea east.

 

Posted (edited)
3 hours ago, Kevin Christensen said:

Paradigm choice always involves deciding "Which problems are more significant to have solved?" 

This is the central factor in all paradigm choices in Pragmatism as a philosophy.  And so, justifiably Kuhn again enters any discussion about paradigms.

I recall that Hawiians also use different terms for "toward the sea", as opposed to "toward the mountains", and other experiential terms for directions as well. Their many sea voyages, sometimes of a thousand or more miles, to a specific island, show that directions based on human experience as opposed to scientific abstractions, like NSEW, work just fine.

And so we have yet another example of how truth can be based on personal experience rather than the universal abstractions of a "scientific" system.

Does Santa need a compass? ;)

Just remember, to get to the grocery store, you have to turn right at Jack-In-The -Box.

Edited by mfbukowski
Posted (edited)
11 hours ago, Kevin Christensen said:

If you want to use animals and modern compass directions your key issues to push Malaysia and discredit Mesoamerica, that is entirely up to you. 

Please read the original post. I'm not using animals and modern directions to push any model, nor am I discrediting Mesoamerica. If you feel the prompts given so far have been somehow biased for a specific model, I'm open to adjust the prompts. However, if I were to go into details in my general prompts about 4th century Mesoamerican spatial references to boost the score given to Mesoamerican model and not provide the same details for the other models, the methodology would be biased for Mesoamerica. Not going to do that at this stage.

I am open to submitting more detailed prompts, but I'm not going to be the best person to write those up for Mesoamerican model. Would you be willing to do so? Do you know someone else that might be willing to do so?

11 hours ago, Kevin Christensen said:

If you don't think Joseph Smith's report of Moroni's comments about "the former inhabitants if this continent" amounts to problem at all, certainly not worth mentioning to Chat GPT, that too is entirely up to you. 

I'm not going to include Joseph Smith's reports of what the angel had said to him about the origins of the book because his comments do not come from the plates. It doesn't come from within the book. The purpose of this exercise is to evaluate the setting based on what is in the text. But let's not forget that you cut off an important part from Moroni's comment. Moroni told Joseph that the book was a record of "the former inhabitants of this continent AND the source from whence they sprang" Where did the former inhabitants of the American continent spring from? The data is clear the indigenous inhabitants of the American continent came from East Asia, Siberia and Australasia. I'm confident Moroni's comment fits the data.

Edited by Zosimus
Posted
11 hours ago, Kevin Christensen said:

I also notice that the only reference to elephants in the Jaredite account (Ether 9:19) is in the 4th generation from arrival (Ether 1:30), which means to me that pushing them to 600 BC for your Chat GPT inquiry is highly questionable.  

I included elephants within the Jaredite timeline because I don't want to start guessing when an animal may or may not have gone extinct since extinction of elephants is not mentioned in the text. But if you have an estimated date range for elephants in the land I can resubmit the prompt. When do you estimate the 4th generation from arrival to have been? If you're proposing elephants went extinct soon after, what's a reasonable amount of time to allow for the extinction?

Posted (edited)

Pending any response from @Kevin Christensen, I'll remind about the open invitation for anybody that feels the prompts are too generic or too detailed to comment with adjustments. Since Kevin commented that it "Seem a bit dubious to ask for a geographical issue by providing zoological information", let's stick to the geography for now. We can address other issues like fauna, flora, metallurgy, weights and measures and DNA later.

Kevin specifically mentioned John Sorenson's and John Clark's internal maps. I agree that's the best place to start. As John Clark said:

"P
roposals for real-world (external) settings for Book of Mormon lands and cities come and go with the regularity of LDS general conferences or market forces, so what was needed was a timeless instrument for judging any geography that may come along—not just assessments of the geographies then in play. The main objective of my essay [is] to outline a key for assessing all external geographies based on information in the Book of Mormon, the ultimate authority on all such matters ... It has been my experience that most members of the Church of Jesus Christ of Latter-day Saints, when confronted with a Book of Mormon geography, worry about the wrong things. Almost invariably the first question that arises is whether the geography fits the archaeology of the proposed area. This should be our second question, the first being whether the geography fits the facts of the Book of Mormon—a question we all can answer without being versed in American archaeology. Only after a given geography reconciles all of the significant geographic details given in the Book of Mormon does the question of archaeological and historical detail merit attention. The Book of Mormon must be the final and most important arbiter in deciding the correctness of a given geography; otherwise we will be forever hostage to the shifting sands of expert opinion." (source)

To assure consistency across this exercise I have created a customized GPT called "Historical Text Explorer". I have added the following configuration:

"Historical Text Explorer is a scholarly GPT specializing in historical texts from 600 BC to 420 AD, with a specific expertise in history, ancient technology, and world geography in the period between 2500 BC and the 5th century AD. It provides detailed, historically accurate analyses, focusing on linguistic, cultural, and geographical context. The GPT offers deep insights into geographical landscapes and their historical significance. More specifically, this GPT will be a tool to evaluate a text published in 1830 that claims to be a history of a Middle Eastern tribe that migrated to an unidentified land in the 6th century BC. This land remained mostly isolated from the Middle East and the Mediterranean world until some time between the 4th century AD and the 15th century AD. Users of the GPT will attempt to identify the land described in the book by analyzing the geographical features described in the book."

For those with ChatGPT+, you can try out the custom GPT here.

I have also uploaded the following documents to the GPT:

  1. Book of Mormon Conceptual Map without labels (source)
  2. John Sorenson's Mormon Map without labels
  3. Anonymized version of the Hagoth to Bountiful section of John Clark's Revisiting “A Key for Evaluating Book of Mormon Geographies” (also attached to this post I. Port Transect to Northern Line Transect..pdf) I will add more as I complete the format conversion.
  4. The schematic below, revised from Figure 2 "Elemental Structure of Book of Mormon Lands." from Clark's Revisiting a "Key for Evaluating Nephite Geographies". To assure the GPT doesn't stumble on preexisting biased info on Book of Mormon geography, I have added the following anonymous labels for Clark's seven transects as follows:


7z7WilS.pngCumorah = Hill Transect
Hagoth = Port Transect
Bountiful = Northern Line Transect
Seashore City = Seashore City Transect
Zarahemla = Central City Transect
Moroni = Eastern City Transect
Nephi = Southern City Transect

Edited by Zosimus
Posted (edited)

To evaluate convergence of the Mesoamerican model with John Clark's internal map, I'll require the assistance of a Mesoamerican model expert. Could someone kindly provide the topogrphical points in the Mesoamerican model that correspond best to the following transects? @Kevin Christensen @Brant Gardner 

Cumorah = (eg. Cerro El Vigia, Mexico)
Hagoth = (eg. San Dionisio Del Mar, Mexico)
Bountiful = (eg. Coatzacoalcos, Mexico)
Seashore City = (eg. Tercer Canton la Curva, Mexico)
Zarahemla = (eg. Santa Rosa, Chiapas, Mexico)
Moroni = (eg. El Bellote, Mexico)
Nephi = (eg. Kaminaljuyú, Guatemala)

If no input from those familiar with the model, I'll proceed with the above examples, taken as best I could see from here, even if it is not ideal

Edited by Zosimus
Posted
On 12/25/2023 at 7:13 PM, Zosimus said:

Please read the original post. I'm not using animals and modern directions to push any model, nor am I discrediting Mesoamerica. If you feel the prompts given so far have been somehow biased for a specific model, I'm open to adjust the prompts. However, if I were to go into details in my general prompts about 4th century Mesoamerican spatial references to boost the score given to Mesoamerican model and not provide the same details for the other models, the methodology would be biased for Mesoamerica. Not going to do that at this stage.

I am open to submitting more detailed prompts, but I'm not going to be the best person to write those up for Mesoamerican model. Would you be willing to do so? Do you know someone else that might be willing to do so?

The prompts and responses are useless as an indication of anything. You are acting like ChatGPT is an objective authority that can be trusted.

Don’t do that.

Posted
2 hours ago, The Nehor said:

The prompts and responses are useless as an indication of anything. You are acting like ChatGPT is an objective authority that can be trusted.

Don’t do that.

For sure, this experiment isn't going to convince anyone of anything. But I wouldn't say its any less useless than say a fireside on Book of Mormon Geography with Wayne May. Or a couple youtubers sitting around giving very general info about the various geography models out there

Posted

Slowly adding data to the GPT. Didn't get as much done over the holidays as expected, but have uploaded the following docs:

hpjUm1q.png
Will also make the contents available to all for review/edits, but need to complete the conversion of Clark's Revisiting “A Key for Evaluating Book of Mormon Geographies” first 

Posted

Have uploaded the following, revised from John Welch's "Charting the Book of Mormon: Visual Aids for Personal Study and Teaching". This will give the GPT some data on the distances between the Defensive Line (Bountiful), the Central City (Zarahemla) and the Southern City (Nephi).

Ten Essential Features of the  Geography described in the Book

Gt84krQ.png

1. A narrow neck (isthmus) separated the land northward from the land southward and was flanked by an east sea and a west sea.
 
2. Northern and Southern lands occupied at least three times as much western coastline as eastern coastline.
 
3. The eastern wilderness was much wider and lower than the western wilderness but not nearly as wide as the southern wilderness.
 
4. The Southern City was in a highland valley; the Central City was in a large river basin.
 
5. A river flowed northward through the large river basin.
 
6. In the area of the Southern City was probably a highland lake of significant size.
 
7. The Central City was surrounded by fortifications.
 
8. The Southern City was three weeks’ travel south from the Central City and near a lake.
 
9. The Defensive City was north of the Central City and near the narrow neck; it was about five days’ travel from the Eastern City and guarded the route to the land northward.
 
10. The Hill Transect was near the eastern sea, not very far north of the Defensive Line Transect.
 

Many accounts of journeys are found in the Book, and approximate distances can be determined from the details in those accounts. For example, one group traveled approximately 21 days from a lake (near the Southern City) to the Central City. Since the group consisted of men, women, children, and flocks, the journey was probably a relatively slow one. Basing his calculations on the rates of travel for pig herders, one scholar concludes that the group most likely traveled about eleven miles a day. If we take into account the twists and turns of the wilderness through which they traveled, the journey was probably around 250 miles (180 miles if they traveled in a straight line, as shown on the map). The distances for the other five journeys listed on this map were determined in a similar fashion.
  • 2 weeks later...
Posted
On 12/23/2023 at 6:33 AM, Tacenda said:

It's easy to see that scripture is muddled with men mingling. 

No, it's clearly the philosophies of men mangled by scripture.

🤭😱🤯😉

Posted (edited)
On 1/3/2024 at 10:16 PM, Zosimus said:

Have uploaded the following, revised from John Welch's "Charting the Book of Mormon: Visual Aids for Personal Study and Teaching". This will give the GPT some data on the distances between the Defensive Line (Bountiful), the Central City (Zarahemla) and the Southern City (Nephi).

Ten Essential Features of the  Geography described in the Book

Gt84krQ.png

1. A narrow neck (isthmus) separated the land northward from the land southward and was flanked by an east sea and a west sea.
 
2. Northern and Southern lands occupied at least three times as much western coastline as eastern coastline.
 
3. The eastern wilderness was much wider and lower than the western wilderness but not nearly as wide as the southern wilderness.
 
4. The Southern City was in a highland valley; the Central City was in a large river basin.
 
5. A river flowed northward through the large river basin.
 
6. In the area of the Southern City was probably a highland lake of significant size.
 
7. The Central City was surrounded by fortifications.
 
8. The Southern City was three weeks’ travel south from the Central City and near a lake.
 
9. The Defensive City was north of the Central City and near the narrow neck; it was about five days’ travel from the Eastern City and guarded the route to the land northward.
 
10. The Hill Transect was near the eastern sea, not very far north of the Defensive Line Transect.
 

Many accounts of journeys are found in the Book, and approximate distances can be determined from the details in those accounts. For example, one group traveled approximately 21 days from a lake (near the Southern City) to the Central City. Since the group consisted of men, women, children, and flocks, the journey was probably a relatively slow one. Basing his calculations on the rates of travel for pig herders, one scholar concludes that the group most likely traveled about eleven miles a day. If we take into account the twists and turns of the wilderness through which they traveled, the journey was probably around 250 miles (180 miles if they traveled in a straight line, as shown on the map). The distances for the other five journeys listed on this map were determined in a similar fashion.

Haven't seen you in a few days -- I hope your uploads didn't give the AI a reason to quit!

7jGf.gif

 

Edited by CV75
Posted
3 hours ago, mfbukowski said:

No, it's clearly the philosophies of men mangled by scripture.

🤭😱🤯😉

That happens sometimes. 🤕😏

Posted (edited)
2 hours ago, CV75 said:

Haven't seen you in a few days -- I hope your uploads didn't give the AI a reason to quit!

Training the GPT has been more challenging than I expected, but I haven't given up and the AI hasn't quit yet. The challengs has been uploading data about the Book of Mormon that isn't obviously from the Book of Mormon, because as soon as AI figures out its from the Book of Mormon it gets all woke. Or rather, once the GPT identifies the text as the Book of Mormon it gets cagey and doesn't want to offend anyone. So the goal here is to give it information about the geography and material culture described in the book in a way that it can't identify the book.

Edited by Zosimus
Posted
2 minutes ago, Zosimus said:

Training the GPT has been more challenging than I expected, but I haven't given up and the AI hasn't quit yet. The challengs has been uploading data about the Book of Mormon that isn't obviously from the Book of Mormon, because as soon as AI figures out its from the Book of Mormon it picks up a bias for existing models due to the sheer volume of research on the Internet.

In other words, once the GPT identifies the text as the Book of Mormon it gets all cagey and doesn't want to offend anyone. So the goal here is to give it information about the geography and material culture described in the book in a way that it can't identify the book.

This is certainly an interesting exercise in how AI works!

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...