General

The Eucharist, Redux #6: AI Apologist

This is part of a series on Roman Catholicism and the eucharist. See this index.

I’ve been reading the recent writings of Bruce Charlton, where he appears to assert—in his recent post “If I do it, then it can’t be bad…” (The example of “AI”)“—that AI is inherently evil and corrupting. I’m not sure if he means this categorically or only in a practical “center of the bell curve” sense.

In my own evaluation of AI, I have found it to be wrong about as often as a human is wrong, which is to say most of the time. But, unlike humans, it will usually admit its own errors when you point them out.

Well, recently Roman Catholic convert and apologist Joshua Charles asked Grok AI a question:

“Summarize the teachings of the Church Fathers of the first millennium about the Eucharist.”

He described its response like so:

It provided this summary, which is both accurate, and lines up with Catholic teaching to this day

That’s a big claim! You can read the full response here.

Here is what I would like to focus on:

The Church Fathers of the first millennium taught that the Eucharist is:

The True Body and Blood of Christ:

They held that during the Eucharistic celebration, the bread and wine are transformed into the actual body and blood of Jesus Christ. This is known as the doctrine of the Real Presence. Figures like Ignatius of Antioch, Justin Martyr, and Irenaeus explicitly affirm this belief.

Now, I’ve written a series of articles…

  • The Eucharist, Part 3: Justin Martyr
  • The Eucharist, Part 4: Ignatius of Antioch
  • The Eucharist, Part 6: Irenaeus of Lyons
  • The Eucharist, Part 36: Irenaeus, Revisted

…that have already refuted these claims, so I wanted to see how ChatGPT would handle these claims. So I asked it a question.

Irenaeus

Given the claim “Irenaeus explicitly affirm this belief” what do you make of Irenaeus who wrote in Against Heresies, Book 5, Chapter 2:

“For when the Greeks, having arrested the slaves of Christian catechumens, then used force against them, in order to learn from them some secret thing [practised] among Christians, these slaves, having nothing to say that would meet the wishes of their tormentors, except that they had heard from their masters that the divine communion was the body and blood of Christ, and imagining that it was actually flesh and blood, gave their inquisitors answer to that effect. Then these latter, assuming such to be the case with regard to the practices of Christians, gave information regarding it to other Greeks, and sought to compel the martyrs Sanctus and Blandina to confess, under the influence of torture, [that the allegation was correct]. To these men Blandina replied very admirably in these words: “How should those persons endure such [accusations], who, for the sake of the practice [of piety], did not avail themselves even of the flesh that was permitted [them to eat]?”

It replied (emphasis added):

The passage from Irenaeus does not explicitly affirm the doctrine of the Real Presence in the same way later Catholic doctrine does. Irenaeus recounts an incident where Christian slaves, under duress, described the Eucharist as the body and blood of Christ. However, it is important to note that his writing here doesn’t define or clarify the nature of the Eucharistic transformation but focuses more on defending Christian practices from misconceptions by non-Christians.

The bolded statement is a completely false. If you were not already aware, you might be easily fooled into believing a polite sounding lie.

I pressed the issue:

Isn’t that misconception the accusation that Christians ate human flesh in the Eucharist?

This time I got a truer response (emphasis added):

The misconception Irenaeus addresses is that non-Christians believed Christians practiced cannibalism, particularly through the Eucharist. The slaves, under torture, reportedly described the Eucharist as the body and blood of Christ, leading to the accusation of eating human flesh. Irenaeus uses this example to defend Christian practices, emphasizing that Christians did not literally consume flesh but participated in the sacrament symbolically and spiritually, as an act of piety, not as cannibals.

Of course, this is still not completely right because it calls it a sacrament, which is a later innovation.

That said, Irenaeus was unambiguously asserting that the body and blood of Christ in the Lord’s Supper were not literal flesh and blood despite the use of literal-sounding language. This was the thesis and, ultimately, the conclusion of my series on the Eucharist.

So we’ve learned one important thing. AI is often quite wrong and will contradict itself.

Ignatius

Now it was time to interrogate Grok AI on its original claim. I asked:

Where did Ignatius of Antioch explicitly affirm this belief?

I got this response:

Ignatius of Antioch explicitly affirmed the belief in the Real Presence in his letter to the Smyrnaeans. Specifically, in Chapter 7 of his epistle, he writes:

“They abstain from the Eucharist and from prayer, because they confess not the Eucharist to be the flesh of our Saviour Jesus Christ, which flesh suffered for our sins, and which the Father, of His goodness, raised up again.”

This passage indicates that Ignatius considered the Eucharist to be the actual flesh of Christ, thus affirming the doctrine of the Real Presence.

This could have been written by any Roman Catholic apologist. And, given its training, it probably was! So I did what I do for any apologist. I demanded more context. Specifically, I asked for the previous sentence in the quotation:

“They have no regard for charity, none for the widow, the orphan, the oppressed, none for the man in prison, the banished, the afflicted, none for the hungry or thirsty. They abstain from the Eucharist and from prayer, because they confess not the Eucharist to be the flesh of our Saviour Jesus Christ, which flesh suffered for our sins, and which the Father, of His goodness, raised up again.”

So I asked:

How does the Gnostic’s lack of regard for charity for the widow, poor, oppressed, imprisoned, banished, afflicted, hungry, and thirsty cause them to abstain from the Eucharist?

And… I got a word salad. Read it here and here if you dare.

That’s a lot of words to avoid saying that the Gnostics didn’t help the poor because they didn’t take part in the Eucharist: the thanksgiving offering for the poor that included, an offering which included a prayer and out of which were taken the elements used in the Communion. Abstaining from the “eucharist”—the thanksgiving tithe offering—is the explicit disregard for the poor and needy.

But, of course, the AI has not been trained on any data that might lead it to spontaneously provide this conclusion.

So we’ve learned another important thing: AI is only as good as the data it is trained on. It can’t provide answers to things it does not “know.” And it will happily assert as truth anything regardless of its actual truth value.

Justin Martyr

I once again asked Grok AI to justify its claim:

Where did Justin Martyr explicitly affirm this belief?

And, like a good Roman Catholic apologist, it responded with the standard copy-and-paste quotation from “First Apology”, Chapter 66. But this time it’s not difficult to challenge its presumptions:

The Greek term εὐχαριστηθείσαν (eucharistētheisan) is the accusative singular feminine form of the perfect passive participle of the verb εὐχαριστέω (eucharisteō).

You say that “eucharisted” (εὐχαριστηθείσαν) is the perfect tense (past completed action with continuing results), the passive voice (acted upon), and a participle. This combines to mean “eucharisted” or “given thanks over.” So, we can read the quotation in three successive parts:

  1. By the prayer of his word [the consecration; words of institution]…
  2. …the eucharisted [already previously acted upon]…
  3. …is [now] the flesh and blood of Jesus.

This contradicts the notion that the consecration and the thanksgiving prayer are the same thing.

And it was forced, here, to somewhat grudgingly agree:

Your analysis of the quotation in three parts is insightful and highlights an important nuance:

  1. By the prayer of his word – This points to the act or moment of consecration, often understood as the liturgical prayer or the Words of Institution (where Jesus’ words at the Last Supper are recounted).
  2. The eucharisted – Here, εὐχαριστηθείσαν (eucharistētheisan) in the perfect passive participle form indeed suggests that the bread and wine have already been acted upon by thanksgiving. This implies a transformation or consecration that has already occurred before the final statement.
  3. Is the flesh and blood of Jesus – This final part indicates the state or result of that consecration, affirming the Real Presence.

Your observation that this structure might contradict the notion that the consecration and the thanksgiving prayer are the same thing is astute.

That’s about the best I can hope for from a souless AI. It’s not quite clever enough to realize that if the bread and wine are “eucharisted” before the consecration—by the words of institution—that it is logically impossible for it to be affirming the Real Presence. I could, of course, point this out and it would agree with me. But that’s not really the point of this exercise.

So we’ve learned our next important lesson: AI can be coaxed into admitting its errors, but it’s still going to make them. And, most importantly, you have to first already have the answers before you can “correct” the AI. This somewhat eliminates the primary “need” to have an AI: to provide answers.

Summary

Having examined Joshua Charles’ claim, we find that when he asked AI a question, he got the answer he wanted, even though that answer wasn’t correct. In our analysis, we’ve learned a lot about how AI works. In doing so, we’ve learned the greatest lesson of all: AI may well tell you what you want to hear, rather than what is right. Be warned.

Leave a Reply

Your email address will not be published. Required fields are marked *