Symbol Preaching in the Digital Age: From Symbol Recognition to Symbol Interpretation in Facebook Ads

Abstract

The thesis of this paper is that in the digital age we are moving away from words and concepts characteristic of the print age, towards the use of images and symbols. I distinguish between objective symbols as in mathematics, and cultural symbols as in poetry and religion. Students must learn to move from recognizing the objective rules of language to internalizing the norms of culture, according to the analogy of learning. Ricoeur’s theory of interpretation explains the passage from recognition to interpretation in the cultural sciences. This passage is not only cognitive but also implies the discovery of an experiential dimension, as in poetry and worship. This theory is applied to the findings from religious ads on Facebook. By creating new audiences by trial and error, the number of viewers increased from 1 K to up to 100 K. The analysis revealed that viewers showed little interest in informational and moralistic ads, but favored symbolic presentations of the Passion, the Resurrection, the Transfiguration, the Eucharist, the origin of evil, etc. The conclusion offers guidelines: the need to advertise, to adapt to audiences, to get feedback, and to preach through symbols rather than concepts.

1. Introduction

The thesis of this paper is that in the digital age we are moving away from words and concepts characteristic of the print age, towards the use of images and symbols. Any word has a clear dictionary definition, but images are open to an infinite number of symbolic interpretations. The image-centered culture of the digital age does not replace the concept-centered world of the print age, but it adds a new dimension to it. In order to pass from one to the other, readers must overcome the one-dimensional world of rationality and discover the multi-dimensional universe of symbols. This will be accomplished by moving from mere symbol recognition characteristic of elementary education to the level of symbol interpretation in art, poetry, and religion.

This paper will proceed in three steps. I will first define the concepts. Next, I will present my findings from religious ads posted on Facebook. Finally, from these findings I will suggest guidelines for the future.

2. Theory of Symbol Recognition and Symbol Interpretation

The word symbol has two basic meanings, depending on the context. In the sciences, symbols function as signs defined by one specific meaning, as in the mathematical formulas. These are objective symbols. In the cultural sciences and religion, symbols have many meanings. They are cultural symbols. Elementary education teaches the recognition of objective symbols like the letters of the alphabet and the structural parts of a word, while higher education introduces students to the cultural symbols of Shakespeare and the arts. Objective symbols are rational creations of the written culture that must be learned. They are objectively defined and recognized universally. Cultural symbols, by contrast, are special to a given culture or field, although some are shared world-wide. Children do not begin with objective knowledge but with images that blur reality and the imagination, with words that are both objective and subjective. Education must first teach the rationality of objective symbols, that is, symbol recognition. In this paper I take the move from symbol recognition to symbol interpretation (or from objective to culture symbols) as the passage from the rationality of the written culture to that of image interpretation which prevails today.

Let us begin with a very elementary question: what is the meaning of meaning? This was the title of the 1923 book by C. K. Ogden and I. A. Richards. They came up with 23 categories of answers, none of which was more important than the others (Ogden and Richards [1923] 1938). Instead of trying to find a single answer, I will turn to two basic schools of thought, the American school of philosophers of language Charles Peirce (1839–1914) and Charles Morris (1901–1979), and the European tradition of semiotics initiated by Ferdinand de Saussure (1857–1913). My presentation is obviously a simplified version of the very diverse field of semiotics.

Morris and Peirce were mainly interested in the psycho-social processes relating objects and words. For them, “something is a sign only because it is interpreted as a sign of something by some interpreter (Noth 1990, p. 49)”. This perspective emphasizes the objective dimension of communication, namely, the speaking subject, the topic of communication, and the language of the written or spoken words. This view has many consequences for peaching. When reading the bible in this perspective, the preacher will inquire about the objective meaning of the words in Greek or Hebrew. When a passage contains a story, the preacher may develop the psychology of the biblical actors. Finally in the applications, the emphasis will be on willful acts and decisions.

Quite different is the semiotic tradition which focuses on linguistic and syntactic structures. Now words are analyzed not just within sentences as in Aristotelian rhetoric, but within a whole cultural corpus, e.g., all the writings of a given author or a literary genre. In exegesis, historical criticism will look for broad scriptural structures, e.g., the four sources of the Pentateuch or the differences between the three synoptics. In all forms of creative learning—and in preaching—the basic structure is that of an analogy between a source and a target domain. Learning involves the analogical imitation of a model, e.g., a literary masterpiece, an artwork, the heroes of history, or the examples found in the scriptures.

These two schools are complementary. Here is my outline for this section. (1) I will first turn to semiotics to define the basic structure of learning and preaching as analogical. (2) Next, I discuss the move from symbol recognition to symbol interpretation according to Ricoeur. Finally, (3) I show that symbols add an experiential dimension to reality, as in psychoanalysis and religion. After this theoretical first part, I present findings from Facebook ads, and conclude with guidelines for preaching in the digital age.

2.1. The Analogical Structure of Learning and Preaching

An analogy is created by the structural similarity between a source and a target domain. Its purpose is to transfer knowledge from one to the other. Hence there are three elements: two structures which must be similar and a subject or mind that sees the similarity between the two (Holyoak and Thagard 1997, pp. 35–44). A mother may expressively say “mama” or “papa” to invite the infant to imitate these sounds. Here we have the model “mama,” the imitation of the model by the child, and the need for the child to understand the relationship between the two. In infancy the child may learn how to use the plural of nouns simply by adding an “s” in imitation of what adults do, but this imitation will be successful only if the child understands the implicit rules of grammar. Over the years the child learns basic rules of social behavior not just grammatical rules. Once these structures are internalized, a teenager can find solutions to new situations, e.g., the plural of a word in a foreign language or the appropriate behavior at college away from home. In elementary school the child is motivated by its desire to please parents, but in high school it may lose interest in learning, and also reject religion and church attendance. Motivation is a key factor in learning.

The lack of learning motivation is a problem not only in education but also in preaching. The description of the life of Jesus as a model to emulate is not enough to motivate the audience to apply this analogy to themselves. Analogical reasoning, which is basic in all forms of learning, is not enough to change the hearts of people. It is the agency of the listeners which allows to move from symbol recognition to symbol interpretation.

Christian preaching consists of presenting paradigmatic events or biblical texts as interpretive models. Before the text of the New Testament was available, the first Christians used events and texts of the Jewish scriptures as types for understanding the message of Christ. Thus, the psalms were scrutinized for images of Jesus Christ. This typological model of exegesis has been followed ever since. But here is the catch: it did not work with Jews who did not relate Jesus to their scriptures. It did not work well with gentiles who did not know the Jewish scriptures. Today it does not work when the biblical model does not fit the listeners’ expectations. The message of Jesus Christ is meaningless to those who doubt the existence of God. Cradle believers may suddenly stop believing when they can no longer make sense out of the religious symbols of their childhood.

2.2. Moving from Symbol Recognition to Symbol Interpretation

In elementary education and catechetical teaching, symbol recognition is usually sufficient. In the first grade, students learn to recognize words ending with -at (as in cat), then -an (as in man), etc. In a fifth-grade social science class, they must recognize “the most common gas on earth’s atmosphere”. And in tenth grade, after reading a one-page letter from Abigail Adams to her husband, they must recognize its basic ideas in a multiple-choice test1. In national exams, multiple-choice testing has become the norm when there is no practical alternative. There are multiple-choice exams even in college, being preferred by both students and teachers. Obviously, there is also symbol interpretation throughout the course of elementary and secondary education, not just symbol recognition.

I follow Ricoeur in his Interpretation Theory (Ricoeur 1976, p. 74) to explain the passage from recognition to interpretation in the cultural sciences. In the humanities and the social sciences (mainly in cultural anthropology, clinical psychology, and qualitative sociology), the mind must go back and forth between a text and the understanding of its meaning. For Ricoeur, interpretation is the process of multiple readings with increased understanding; interpretation may never be complete. The first reading may be naive, the second one a little more sophisticated. At the beginning, understanding is a broad guess of the general meaning. At the end, we reach appropriation. The latter is a “dynamic of interpretative reading;” it does not lead to a transfer of knowledge from the speaker or writer to the listener; it is much more. “Interpretation is completed as appropriation when reading yields something like an event, an event of discourse” (Ibid., p. 92). Interpretive reading is the creative process which produces a discursive event in the mind of the reader, says Ricoeur. It is not a thing, but a new perspective, that of a fusion of horizons. In appropriation, “the world horizon of the reader is fused with the world horizon of the writer” (Ibid., p. 93). It is not a transfer of knowledge but a sharing of perspectives.

For Ricoeur, interpretative reading is a process like what I call symbol interpretation. Appropriation does not mean taking possession of the message of a writer; it means accepting “a mode of being in the world that the text opens up.” When the meaning of a text has been appropriated, the reader has been transformed. For Ricoeur, interpretative reading—or symbol interpretation—reveals “new modes of being [which] give to the subject a new capacity of knowing himself”. Ricoeur gives as an example of appropriating the letters of Paul which are read and explained in Sunday worship. When the bible is read interpretatively, it is transformative. Unfortunately, not many people reach this high level of understanding.

In Ricoeur’s perspective, one moves from the recognition of objective symbols to their interpretation through multiple readings. It is this repeated reading which is the bridge between the two structures of an analogy, the knowledge of objective symbols on the one hand and the perception of meaning on the other. In fashion, each individual bridges the gap between the objective models of fashion and personal preferences. In biblical understanding, one moves from the objective understanding of a biblical text to its understanding through multiple readings in a deliberate and active pursuit. At a global level, one will move from objective symbols to their interpretation through a deliberate and active pursuit of meaning in all areas of knowledge. For this, the use of symbols as cultural creations adds a new dimension. Moving from objective symbols to symbol interpretation mut also take place in preaching. How this move can be achieved will be discussed through examples from Facebook ads below.

2.3. Cultural Symbols as Experience of a New Reality

When words are used as cultural symbols, they add a new dimension, that of depth and experience. The phrase, “A woman with a mysterious smile” is a one-dimensional discourse. Leonardo’s painting of Mona Lisa is two-dimensional: an image and a message. No number of words can express this mysterious smile, but art and poetry can. Looking at the Alpine scenery of snow-capped mountains, cows only see grass. When one is blind to the symbolic dimension, words are only things; they describe a world without art, a scenery of grass without the beauty of snow-capped mountains. Any objective symbol can be seen symbolically, but it requires a qualitative jump.

Symbols in psychoanalysis and religion belong to two universes; one is linguistic (the story of a dream or a sacred text) and the other is non-linguistic and non-semantic (the hidden psychic conflicts or the “wholly other” in Rudolph Otto’s description of the holy (Otto [1917] 1970)). The second dimension refers to an experience that affects the self, not just the brain. But there are not two significations, one (grass) added to the other (the Alpine scenery), because an implicit dynamic must unite the two.

According to Ricoeur, discursive language (that of the written culture) belongs to the order of logos (reason) while symbols belong to bios (experience). What complicates things is that the experiences of bios are usually expressed according to the linguistic rules of the logos, (Ricoeur 1976, p. 59) that is, in words rather than in images. How can we move from words to experience?

The distinction between cultural symbols and ordinary language may help our understanding of symbols as experience. I can describe a dream to a friend without understanding its psychoanalytic meaning; I just tell the facts. I can write a sincere love letter but without being able to convey my deep feelings symbolically; I just give information. I can quote the bible like a newspaper; it may show no faith. Ordinary language belongs to the order of the logos, of objective symbols that describe objective realities; the listeners may get the intellectual message but not the depth of the experience. But psychic conflicts, deep feelings, and mystical experiences are lived experiences that transcend ordinary language. Knowing the technical language of specialists is not enough. One may assemble a collection of technical terms as Freud did in the Interpretation of Dreams or Mircea Eliade in Patterns in Comparative Religions (Eliade 1965), but knowing that a wolf is a psychanalytic symbol, and a mountain top is a symbol of divine encounter does not make me a psychoanalyst or a religious prophet. These words become symbols only, for instance, when wolf is experienced in a dream as a threat, and mountain top is experienced as an encounter with the divine, e.g., in meditation.

The interpretation of texts through multiple readings belongs to the age of print. Symbols are more prominent in the digital age. Religious teaching is usually just the explanation of doctrines or sacred texts, and so is explanatory preaching. More appropriate for our age is symbolic interpretation and preaching as the sharing of experience. This is what happens in successful religious advertising.

3. Findings from Facebook Ads

Sixty-nine percent of Americans use Facebook. Seventy-three percent of U.S. adults visit Facebook every day; 37% of them get their news from Facebook; most Facebook users access the app on a mobile phone (Carmicheal, no date) (Carmicheal 2022). In the field of popular entertainment, the top ten world celebrities receive over 100 M visitors every day; Lady Gaga (the only name I could recognize) ranked 128th with 55 M visitors daily. There are probably more visitors to celebrities in a single day than to all US churches in a month. The social media function as constant social reinforcement: Americans check their cell phones every ten minutes and the 18- to 24-year-olds twice as often, and on average 96 times a day (Williams 2021). By contrast, they go to church—if they do—only four times a month. Hence it seems imperative that the gospel be present in this environment.

Now some personal data. In 2016, I started posting weekly biblical reflections on the internet without ever checking how many people read them—actually, very few. Realizing that advertisements on Facebook were available for as little as 5 USD a day, I started advertising in November 2019 for USD 7.00 a day in a three-day week (Friday to Sunday or USD 21 a week). I have not increased this amount since then.

3.1. Methodology and Research Design

In social sciences, a research project usually involves (1) a sample, (2) several independent variables like the age, religion, education, and income of the respondents and (3) one dependent variable, in my case the number of people who see my ads and their approval or engagement rate. Such a research design is not possible on Facebook.

One cannot select a sample, only the amount of USD one wants to spend, and the country where one wants to advertise. The number of people who will see the ads is decided by the Facebook algorithm. For each ad, the computer evaluates its desirability for the viewers. Thus, my ad of 5 November 2022, was seen by 118.8 K viewers in the English-speaking audience of about 8.5 million people, and by 328.8 K viewers in the French Speaking audience of only 2.5 million. The difference in the number of viewers is mainly decided by the computer and I cannot make sense out of it. Moreover, if the same ad were advertised every week for six months, the results would not be different because each week the selected viewers would be different; they are not selected randomly as in academic research.

The age, religion, education, and income of the respondents are not known. I do not know the country of the respondents. Thus, on 15 January 2023, I posted an ad to be seen in all the countries of the world and in all possible languages. It was seen by 169 K people, but I do not know their country or language; actually, I know nothing about them.

The independent variable cannot be clearly defined. Is it the number of viewers (which is said to be an estimate?) Or the number of people who click “like” or “love?” Or those who write a comment? To make things worse, for these measurements Facebook gives two sets of statistics, and I cannot make sense of their differences; in any case, no one gets to see the computer language which measures these variables.

This system works to the satisfaction of all advertisers because it increases sales, often considerably. I also find the results trustworthy and reliable because I always used the same method and selected the same statistics, while also introducing limited changes. This is what I have done for the data presented below. In short, this was not a research project, not possible on Facebook, but an experiment which is described in its various phases.

My advertising has gone through many phases. First, I experimented with many audiences to increase the number of viewers. Then I somewhat changed the content. Here is my outline for this section: 1. How audiences are selected on FB. 2. My attempts to improve my audience. 3. Presentation of basic findings. 4. Moving from religious information to faith experiences. 5. Getting involved through comments.

3.2. The Problem of Audience Selection

Audience selection. There is a Facebook tool to help the selection of an audience according to people’s preferences; but this option has not been available anymore since mid-2022. Thus, in 2019 one could ask how many people are interested, e.g., in tennis, a car model, or Protestantism. This tool indicated the number of interested people, their gender, and their ten basic preferences, namely their preferred retail company, their preferred public speaker, political candidate, news media etc. When in December 2021 I selected “Protestantism” as my U.S. audience, the computer indicated that this audience was 74% female; Walmart was the preferred retailer for 28% of them, Donald Trump was the preferred political candidate for 21%, and Fox News the preferred source of information for 18%. When I selected “Catholicism,” the audience was 78% female; Franklin Graham was their public figure (for 29%) although he was not a Catholic; and Donald Trump was their preferred political candidate. It is important to note that people interested in Protestantism or Catholicism are not necessarily Protestant or Catholic, as we do not know their religious identity, only their religious preferences.

Sample selection is a major concern in advertising. On Facebook (henceforth FB) one cannot select a sample of a given size consisting of people of specific social backgrounds. Instead, one can only select people with specific interests. The FB mega-computer codes the interests of all FB members expressed in their postings. As a consequence, the FB algorithm can find appropriate viewers for nearly any kind of ad.

In 2019 I wanted an audience interested in biblical reflections about the Sunday readings of the Catholic lectionary. I wanted an audience that would be ideologically neutral, neither predominantly conservative nor progressive. I selected three available interests: Catholic devotions (conservative), charismatic renewal (progressive), and Pope Francis (progressive in most parts of the world but not for conservative Americans). These criteria yielded a U.S. audience of 7 to 8 million viewers that was 76% female and mainly conservative. With this audience the response rate to my ads was low. I tried similar audiences in Canada, Ireland, and the United Kingdom with similar modest results.

In July 2020 I expanded by including India and the Philippines. Then the results increased ten- to twenty-fold. When the audience included the five English speaking countries of India, the Philippines, Nigeria, Tanzania, and Ruanda, the number of responses from America became insignificant. Consequently, I limited my audience to the five foreign countries just mentioned. While the American audience was 76% female and old, that of these five foreign countries is about 70% male and young, and their response rate is about ten to twenty times higher.

The FB algorithm first evaluates a submitted ad and then sends it to appropriate recipients. The viewers can react to a post by clicking “like,” or “love,” or “share” or write a comment. The number of people who react is the engagement rate. This is the most important statistic: it indicates the importance of an ad to the viewers.

The number of viewers of an ad is decided by the FB algorithm. My top three ads in 2021 were seen by a weekly average of 58.9 K viewers while the three least popular ones were seen by 16.9 K people. The advertisers do not have any input about the number of viewers, but they can change or increase their target audiences.

3.3. Selecting a Better Audience

My main goal in advertising biblical reflections was at first to increase the number of viewers. It is only later that I paid attention to the quality of the responses. There are no techniques for audience improvement. It is by accident that I discovered that by changing my audience, I reached more viewers. When at the beginning I got 1000 or more viewers in my American sample, I had good reasons to be satisfied. Adding Canada, Ireland, and the U.K. made no difference, but adding the Philippines increased the number of viewers substantially. This prompted me to add India. Later I added English-speaking countries in Africa. Finally, I created a French audience (France and 11 African countries) using the English ads in a translation. Here are the various steps of my audience development.

• An audience of 60 M in the US. About 1000 to 3000 viewers. From 9/2019 to 5/2020.
• Audience of 50 M in the US, Canada, Ireland, Philippines. About 10 to 20 K viewers. From June 8/2020 to 10/2020.
• Audience of 650 M of all the FB viewers in US Canada, Ireland, Philippines, India, Nigeria, Tanzania. About 30 to 40 K viewers. From 10/2020 to 5/2021
• Audience 500 M from Philippines, India, Nigeria, Tanzania (without the US). About 40 to 60 K viewers. From 9/2021 to 6/2022.
• Lookalike audience of 8.5 M from Canada, U.S., India, the Philippines, Nigeria, Tanzania, Rwanda. About 80 to over 100 K viewers. From 6/2022 to 9/2022.
• French version of the English ads. Audience of 2.5 M with about 80 to over 100 K viewers from 5/2022 to 9/2022.

There are no recipes for increasing one’s audience; it is mainly a question of trial and error. I do not know why my recent audience of 8.5 M got better results than the previous ones of 650 M. I used a FB tool to create “lookalike” audiences, that is, new audiences with the characteristics of previous successful ads, that is, national audiences that had the characteristics of people who liked my ads. What these special characteristics are, and how lookalike audiences are created is the secret of the computer algorithm. In trial- and error- learning, what counts are the results. The French audience of 2.5 M is much smaller than all the English ones. At first, its performance was worse, but progressively it improved and at times surpassed the English results. Audience selections may be unpredictable, but in the long run, trying new methods is likely to produce positive results; innovation is the key. The same is likely to be the case in preaching.

3.4. Presentation of the Basic Findings

A FB ad consists of a picture and three to five lines of text. The viewers can react by clicking “like,” “love,” “share” or write a comment. The number of reactions per ad is called the engagement rate. Here are the ads from 2020 and the first half of 2021 that received the lowest approval rates. These rates are high in comparison to the average for all ads on FB which is supposedly 0.03%. The ads below are about 15 times higher than this average of 0.3%. Here are the titles:

9.8% Golden-rule Christians are kind and helpful.

10% Three options for after-pandemic church life.

10.2% Most addictions are pleasant.

14% Nature is the image of God.

14.3% Mary’s Assumption is also called her Dormition.

16% Jesus’s position was inclusive: “Who is not against us is for us.”

16.8% All things come from God and return to him.

17.7% No need to shout: God is not deaf.

18.7% The first Christians shared meals and possessions.

19.3% When life is like jumping from a cliff, we need faith.

20.6% The 2021 readings are from Mark.

21.2% Pentecost: Unity in diversity.

Here are a few more titles of the same type as those above:

Simon was a fisherman living next to the lake. His house has been excavated.

We are pilgrims: all flesh is like grass.

The opposition between flesh and spirit is basic in our spiritual struggles.

We live in eschatological times.

Conflicts between liberals and conservatives have existed since the beginning of the church.

Worldwide migration is likely to increase due to climate change.

There are checks and balances in the Catholic Church.

We are in a coronavirus winter. We must become self-reliant.

The content of these ads is informational, cognitive, and moralistic. They present information for symbols recognition, not symbol interpretation. Some are only informational: the Assumption is also called the Dormition, or the 2021 readings will be taken from the gospel of Mark. Many are general: nature is an image of God, all things return to God, the first Christians shared possessions, Jesus’s position was inclusive, and Pentecost was unity in diversity. Many ads are moralistic: addictions are pleasant, God is not deaf, we need faith, prepare for Lent. The lowest rating went to “The Golden Rule Christians.” This title refers to a sociological paper by Nancy Ammerman2 showing that the lived religion of Americans is centered on morality not doctrine.

Like any Sunday preacher, I tried my best, not getting any feedback. Moreover, the above topics, in my view, are like what can be heard in many churches.

For months I was satisfied. It is only when reviewing my results for a paper that I was struck by the following results which I had not noticed. Here are the ads with the highest engagement rates from 2021.

57% Christ is risen.

56% Christ is risen (another week).

53% God is beyond our images.

52% God reveals himself in people’s lives.

49% A common false belief: God punishes you for your sins.

48% Both awe and joy at the Transfiguration.

48% Proclaim the good news to the end of the world.

47% Christ is really present at Mass.

47% What is salvation?

47% What is the origin of evil?

46% Yes, God still heals today.

45% The Passion of Jesus Christ.

43% The Passion of Jesus Christ (another week).

44% The Easter joy will last forever in the eternal banquet.

The above examples suggest that people wanted to learn about the mysteries of faith and the core insights of Christianity. This is also what people in the pews probably expect from preachers.

What can explain the change from low engagement in 2020 to high engagement in 2021? How can viewers (and the people in the pews) reach the level of symbol interpretation when the writer or preacher communicates at the level of symbol recognition, that is, the level of information and morality? It seems that there must be first a transformation in the writer or preacher, rather in the viewers or auditors.

3.5. Moving from Religious Information to Faith Experiences

Starting in the middle of 2021, I changed my writing procedure. In the ads with low engagement rate, I first reflected on the text, and it is only after writing my reflections that I sought a picture to illustrate it. In my new procedure I would first search for a picture to illustrate a main idea of the Sunday reading, and next add the title within the picture (e.g., Jesus is coming soon). The writing of a reflection came third. Now the title and the writing were inspired by the picture selected for its expression of the reading. Previously the process was a dichotomy, my biblical reflections on the one hand and a picture to illustrate my reflections on the other. Now the process was reciprocal, the image giving depth to the understanding of the text and vice versa. Previously I started with a concept from the biblical text, that is, symbols of information; now I start with images as symbols of interpretation.

Moreover, while previously I limited the size of my pictures to about 30 percent of the width of the page, in 2022 I extended the size to 90 percent of the width, which gave pictures a prominent place. What affected the viewers, the text, or the image? The answer is probably both, through a single image-text production. The unity of picture and text is likely to foster active interpretation rather than passive recognition. As explained above, psychoanalytic or mystagogical symbols belong to two orders of things, one is linguistic or iconic, and the other transcendent and experiential. But when the two dimensions are interrelated, the viewer can easily pass from one to the other. In my first series of ads, I juxtaposed a text and an image. I was only communicating at the cognitive and informational level. Now I was personally engaged in the experience of image-text. More generally, while previously I endeavored to give information and moral conclusions, now I tried to explain the mysteries of faith, that is, write about mystagogy.

3.6. The Comments of Viewers

Viewers can click “like” and “love,” but they also write a comment which is a more demanding engagement. Every week about one or several hundred viewers leave a comment. Their number is an indication of the importance of the ad to them. Here are the ads ranked by the number of comments:

579 Sermon on the Mount. “Blessed are the poor.”

502 The Transfiguration is a pre-figuration of the resurrection of Jesus and ours.

477 “Peace be with you!” God’s peace is more than the absence of conflict.

426 “Simon, do you love me?”

397 The return of the lost son. More joy for 1 conversion than for 99 righteous.

375 Blessed are the poor. Blessed are the poor because God cares for them.

372 The four temptations of Christ. “I will give you power and glory if...”

369 Christ King. “Be this sign [the cross] you shall conquer!”

339 “I am with you to the end of time.” “I will send you an advocate.”

321 Martha & Mary. “Martha, you worry about details!”

316 The abundance of wine in Cana was a symbol of the eternal feast.

283 “Remember me in your kingdom.” For the sake of your sorrowful passion, have mercy on us.

275 Our Father in Heaven. “Ask and you will be given.”

261 “He is alive!” “John saw and believed.”

256 “Where is your faith?” Faith gives peace in the middle of the storm.

243 No one can serve two masters. Who is your master: God or money?

237 The friends of Jesus: the outcasts. The prayer of the Pharisee: “God, I do not need you. I am fine.”

224 Jesus is rejected from Nazareth. God is often ignored because people are too busy.

224 Baptism: a passage to a new life.

206 The Good Shepherd. Where is the good shepherd today?

This list gives priority to core beliefs rather than peripheral ones, as in the ranking by engagement rates, but here there is a more personal dimension. People responded more to ads that involved them personally, as in Peace be with you, Simon, do you love me? The return of the lost son. Blessed are the poor. “I am with you to the end of time.” “Martha, you worry about details.” “Remember me in your kingdom.” “Ask and you will be given.” Faith gives peace in the middle of the storm.

The content of these comments suggests that the viewers value faith more than doctrine. At the end of a Sunday service, one may say to the preacher some like, “Well done!” or “I like your special this or that.” Not here. There was never a complimentary comment, but all are affirmations of faith. For instance: Amen, praise the Lord. Praise you Jesus. Amen in Jesus’ name. Jesus Christ is the Lord God. I love you, Lord Jesus. Amen, joy, Jesus. Most of comments simply repeat Amen or Alleluia or both with exclamation points. These simple words are the expression of faith of viewers with low literary skills from the emerging economies of Asia and Africa.

4. Conclusions: Guidelines for Preaching in the Digital Age

I use the word “mystagogy” as the process of leading people to the mysteries of faith, although at the time of Augustine this word referred to the sacred mysteries themselves (and it does so today in the Catholic preparation of adults for baptism). Teaching doctrine is not enough today; preachers must lead listeners to spiritual awakenings. Religious doctrines are intellectual products that can be taught to children in age-appropriate language. Doctrine speaks to reason, and their applications in sermons relate to will power. Teaching and lecturing want to impart knowledge and memorization through symbol recognition. This is useful at all levels, especially the level of beginners. Mystagogical preaching endeavors to transform through the symbolic interpretation of the mysteries of faith. Teaching belongs to the order of the logos, but mystagogical preaching to that of bios in Ricoeur, or to the orders of pathos and ethos in Aristotle. Both types are needed but have different effects, one is more informational, the other more transformative.

My FB ads consisted of over 400 weekly campaigns involving over 5 million people. The bulk of this information leads me to three basic guidelines. Effective preaching in the digital age implies (1) paid advertisement, (2) sermon evaluation, and (3) mystagogy (symbol interpretation) rather than basic religious information (symbol recognition). Here are more detailed guidelines.

4.1. Advertising

Streaming Sunday services and sermons on You Tube is quite common. Unless these events are advertised, few people will know about them. Advertising on FB is inexpensive: a one-day ad is available for as little as USD 5.00 per day. All churches can afford this. Advertising is not enough; it is necessary to check the results. FB, Google and other platforms offer statistics about the number of people who visited a given web page. These programs are free of charge.

The need to advertise has been advocated and practiced by the successful media preachers of the past. Robert H Schuller was known to millions of Americans for his “Hour of Power.” He started modesty by preaching in an open-air movie theater. “Immediately after our first Sunday in the drive-in theater [in March 1955], I sent out a postcard to everybody on the mailing list.” (Schuller 1975, p. 145) It increased his success. Six months later, he invested USD 4000 in advertising, and collected over USD 8000 in donations, which he reinvested in more ads. In Schuller’s opinion, the senior pastor should be personally in charge of advertising. “By being personally responsible for the publicity, I was forced to create, produce, and generate sermons and programs that were newsworthy” (Ibid., p. 146). This was also my experience: writing, posting, and reviewing my weekly reflections forced me to make numerous improvements.

Rick Warren started Saddleback church by going door to door to listen to people’s complaints, and sending a letter to 15,000 homes inviting them to attend his first Sunday service (Warren 1995, p. 139). For him, money and evangelism go together. “Money spent on evangelism is never an expense, it’s always an investment.” But this money is often the first to be slashed. “When finances get tight in a church, often the first thing cut is the evangelism and the advertising budget. That is the last thing you should cut.” He saw both as “the source of new blood and new life for your church (Ibid., p. 202). Advertising is as important as evangelization; usually one goes with the other.

4.2. Sermon Evaluation and Adaptation to Audiences

The performances of teachers, salespersons, and most public agents are evaluated regularly, but sermons are not. FB ads give reactions and comments. Digital technology can provide unique tools of evaluation. For instance, at church during the sermon people in the pews could interact with the preacher by sending text messages on their cell phone. Such an initiative would require overcoming the view of the pastor is the main or sole source of theological learning. Zoom conferences already offer such instantaneous interactions; while the screen shows the speaker, the audience can react by writing messages off screen in the chat room. The speaker can later review these messages and respond if appropriate. Without sermon evaluation, there will be little improvement in preaching.

Any church involves multiple audiences but usually the same sermon is given to each of them (e.g., the Anglos and the Latinos in a given church). Bill Hybels popularized seeker-friendly services that cater to an average customer called Harry. Every aspect of the service (music, topic, language, stories, exhortations) was geared to Harry (Pritchard 1996, chp. 9). Rick Warren developed his own style of seeker-sensitive services by adopting pop music, a church campus built like a shopping mall, and non-denominational sermons preached in street clothes. But what worked for Hybels and Warren in the 1970s and 1980s may not be appropriate in the digital age of today.

The two basic audiences identified by apostle Paul were those living in the spirit and those in the flesh. He catered to both. He was angered by the Corinthians whom he saw as fleshly (1 Co 3:1-3); he also addressed the spiritually mature in most of his letters (e.g., 1 Co 2:6). We all need to improve, at the levels of both beginners and the advanced, At Saddleback, there is a Sunday service for the Harrys and a mid-week service for the advanced. In small churches where there is only one service, the preacher should address both publics, the Harrys and the advanced, but at different Sundays.

Traditionally the members of the two public were identified as the regular Sunday attenders and the devout of pious societies. Today spiritual growth is often gaged in terms of discipleship and a personal relationship with God (Hawkins and Parkinson 2011, chp. 4). Without at least occasional preaching at this advanced level, the spiritual quality of sermons and of the preacher’s life is likely to remain average.

4.3. Preaching as the Interactive Sharing of Spiritual Experiences

“What we have seen and heard, we now proclaim to you, so that our joy may be complete.” (1 Jn 1:3) John did not write to give information about what he saw and heard but to give a testimony of the power of God which he experienced. This is what makes preaching effective: the faith experience of the preacher speaking to the faith experiences of the listeners. Then preaching is at the level of bios (experience) rather than that of logos (rationality and information), although one does not exclude the other.

My FB experience suggests how to move from symbol recognition to symbol interaction. In information preaching, the sermon presents the preacher’s knowledge and wants the audience to accept it. In symbol appreciation, the preacher selects images that evoke faith experiences in him or her, and shares these visual faith experiences with the public; most of the time it works very effectively. Preaching becomes the sharing of faith rather than a sharing of information. In other words, instead of giving objective information about the sightseeing in town, the preacher indicates the one he/she found most attractive. This is required by the very nature of images as cultural symbols. For the merchant, the value of art is in its objective price; for the artist, art is in the eye of the beholder. In the digital age, preachers must be the artists of the word of God.