Biodynamic Viticulture Representations in the French Wine Industry: A Textual Analysis

Abstract

At a time marked by a transition in winegrowing methods, the decision to employ biodynamic viticulture, and its holistic approach, is a choice made by many technicians in the French wine industry. Nevertheless, this practice is perceived very differently among industry members. Some frequently debate its foundations and tangible benefits, while others question the value of scientific research conducted on its methods. The representations motivating the decision to pursue biodynamic practices remain unclear. Thirty members of the industry were interviewed about their opinions and knowledge on biodynamics through semi-structured interviews. A textual analysis of their responses, using Alceste software (Image Ltd., Toulouse, France, version 2018), revealed four key representations of this growing method within the industry. Some practitioners believe that biodynamic preparations may have a material effect on plant behaviour and that the effects of biodynamic viticulture on the final quality of wine should be studied more closely. Others suggest that the anthroposophical foundations of the method should be set aside to allow for its reinvention in a forward-looking manner. Lastly, experimental methods and peer-exchanges are considered essential to understanding the effects of this cultivation method in specific contexts and terroirs. These different viewpoints should be integrated to develop innovative and interesting applications for biodynamic farming methods.

1. Introduction

The foundations of biodynamic agriculture can be attributed to Rudolf Steiner (1861–1925), an Austrian philosopher and the father of anthroposophy. In 1924, Steiner delivered a series of eight lectures to a group of farmers, aimed at addressing the challenges of yield loss, crop diseases, and other issues that had emerged in inter-war agriculture, which was increasingly influenced by the use of chemical fertilisers and synthetic pesticides [1]. Although agriculture comprised only a minor part of Steiner’s work, these lectures were pivotal in fostering a new perspective on farming. Steiner introduced a vision in which the health of soil, plants, and animals is intertwined with the cosmic forces that impart rhythm and structure to their existence. He compared the farm to a living organism, where all components must interact and be set in motion by an impetus, the conscious will of the farmer. He emphasised the significance of practical field trials. Steiner’s lectures had a considerable impact on the emergence of organic farming, and its vision aligned with the principles of agroecology.

Following Steiner’s lectures, farmers and practitioners gathered to exchange their experiences and promote the methods they had learned. They established common standards and practices, fostering a community dedicated to biodynamic agriculture. Consequently, a cooperative named Demeter was established in Germany to process and market products from biodynamic farms. It was not until 1932 that the Demeter Association (Demeter Wirtschaftsverband) was formed to certify biodynamic products. Although the Demeter Association was dissolved during the Second World War but was revived in 1946 by the Anthroposophical Farmers’ Research Circle [2], which subsequently created a brand that was trademarked in Europe, securing its intellectual property rights in 1954. In 1997, Demeter International was established and merged with the International Biodynamic Association in 2020 to create the Biodynamic Federation Demeter International (FBDI). This federation defines the international standards for the Demeter label, a certification mark for products that are produced according to biodynamic farming principles [3].

In France, the Demeter label was recognised by the Ministry of Agriculture in 1982, preceding the Organic Farming label, which was introduced in 1981 and officially recognised in 1991. Originally created for food products, cosmetics, and textiles, the Demeter label quickly expanded to include regulations for viticulture and oenology [2]. In 1995, a separate label specifically for viticultural and oenological practices was established by a group of winegrowers practising biodynamic methods, known as the Biodyvin label [4].

Regarding viticulture, both labels require organic practices in the vineyard (prohibiting synthetic pesticides and fertilisers); however, the Demeter label is stricter than the Biodyvin label concerning the permissible amount of copper used in the vineyard to protect crops, with limits set at 3 kg/ha/year for Demeter and 4 kg/ha/year for Biodyvin. Both labels require the use of biodynamic preparations 500 and 501, and compost preparations derived from seven plants (nettles, horsetail, dandelion, chamomile, valerian, yarrow, and oak bark), either in compost form, as Maria Thun compost, or 500P [5]. Concerning wine production, both labels stipulate that alcoholic fermentation must occur using the yeasts naturally present on the grapes and in the cellar. The total sulphur content in the finished wine is also regulated: for a dry red wine aged for over nine months, Demeter permits a maximum of 70 mg/L of total SO₂, while Biodyvin allows 110 mg/L [6,7]. The requirements of both labels may seem restrictive in relation to Steiner’s original vision. They can be challenging to adhere to under conditions of significant mildew pressure in vineyards or in estates producing barrel-aged wines.

Rudolf Steiner intended for the method he described to be widely adopted, “for everybody, for all farmers” [1]. For nearly a century, Steiner’s philosophy has been largely overlooked. It was only in the early 2000s that a handful of farmers “rediscovered” biodynamics. According to Paull and Hennig, biodynamic practices were present in only 55 of the 183 countries practising organic farming in 2020, representing approximately 30% of these nations [8]. In 2020, Europe remained the continent with the highest prevalence of biodynamic farming, with a significant presence in Germany (34% of global biodynamic practices) and France (6%). Australia ranked second, accounting for 20% of the total biodynamic surface area. These figures only represent certified biodynamic practices, but many farmers employ biodynamic methods without seeking official recognition [8].

In terms of scientific research, there are currently few scientific, peer-reviewed articles focusing on biodynamic practices. Taking biodynamic viticulture as an example, approximately fifty articles investigated biodynamics and its effects on soil, plants, and wine quality. A dozen of these articles were published in the 1990s, with the majority emerging during the 2010–2020 decade. Research in this area is gradually expanding.

Some research associations conduct trials and compile results on biodynamic agriculture, but very few implement controls, resulting in their findings being largely disregarded by the scientific community. For instance, the DOK trial, initiated in Switzerland in 1978 by the FIBL (research institute of organic Agriculture), compared biodynamic, organo-biological, and integrated farming systems by reproducing these methods in a microplot setup [9]. In France, notable research associations include Biodynamie Recherche [10] and Soin de la Terre [11], among others.

Biodynamic agriculture suffers from its anthroposophical heritage. According to Alexandre Grandjean, 39 out of 40 Swiss practitioners do not claim to belong to the biodynamic movement, which is often perceived as spiritual or religious. They are familiar with Steiner’s lectures and principles in a fragmented manner, utilising the biodynamic pharmacopoeia primarily as a foundation for holistic farming practices centred on plant and soil health. These practices are believed to yield more natural crops that reflect their terroir [12].

The French wine industry is currently undergoing a transition marked by significant commitments to environmental sustainability and social acceptance, which involve reducing the use of the most toxic pesticides and replacing them with natural alternatives in the context of high sanitary pressure [13], enhancing biodiversity in vineyards where monoculture has become prevalent [14], lowering the carbon footprint by 2050 [15], and exploring methods for adapting to climate change [16]. Consequently, winegrowers must identify innovative technical pathways to address these requirements.

In this context, a variety of tools are available to farmers, ranging from precision farming to regenerative agriculture, agroforestry, or permaculture; however, choosing among these options is not easy. These decisions will depend on the farmer’s perspective, culture, and history. Rigolot draws a parallel between the worldviews articulated in the literature and the visions of farmers. He argues that it is crucial to understand one’s own worldview, as well as that of others, in order to integrate diverse perspectives and facilitate a shared transformation [17].

Initiated by Durkheim, the concept of individual representations has been extensively employed by Foucault to describe the tools through which we “perceive and understand” the world. Shared individual representations give rise to social representations, which form the foundational basis of cultures. Studying these representations aids in understanding human behaviour [18].

What are the different representations of biodynamics within the wine industry? Are these representations influenced by the wine-growing region or the profiles of the practitioners? How might they affect the industry and the decisions of growers to adopt or reject the biodynamic method? These are the questions addressed in this paper.

2. Materials and Methods

Biodynamics has two distinctive characteristics. Firstly, it is often a broad and empirical practice, making it difficult to establish a specific protocol, identify the effects to observe, and find elements that can be reliably replicated across different contexts. Secondly, it is a topic that provokes considerable discussion and debate within the industry.

We believed that engaging in a dialogue with a panel of viticulturists could provide valuable insights into the biological effects that researchers need to explore. Furthermore, this dialogue would enhance the understanding of the diverse perspectives on the method, fostering more constructive discussions among winegrowers.

2.1. Number of Interviewees

Determining the appropriate number of interviewees for statistical robustness in a qualitative study is challenging. Numerous methods exist for estimating the minimum sample size which depend on the type of study, the nature of the and the dynamics of the anticipated dataset.

For example, scholars such as Janice M. Morse have advocated for selecting a suitable number of participants to reach saturation—the point at which no new information or themes emerge from the data [19]. According to Morse, this number depends on various factors: the study’s objectives, the nature of the subject, the study design, and the quality of the data collected. If the objectives are narrow, the subject is clear to the interviewees, and the panel is successfully selected to yield quality data, the number of participants can be reduced to between 6 and 10 interviewees. If these conditions are not met, it is advisable to increase the number of participants to between 20 and 30. When the amount of information obtained per participant is minimal, interviewing 30 to 60 participants is preferable [20]. Other authors, like Greg Guest suggested that in a relatively homogeneous group, important themes and insights can emerge early in the data collection process and 12 interviews can suffice to “understand common perceptions and experiences among a group of relatively homogeneous individuals” [21].

In our case, we interviewed 30 individuals. Given that our questions are structured, our objectives are clearly defined, and our interviewees possess relevant experience, we consider this number sufficient to provide statistical robustness while remaining manageable for completion within a year.

2.2. Recruitment of Participants and Informed Consent

Participants were recruited through announcements sent via email to professionals across several wine-growing regions. No compensation was provided for their participation. The information collected respects participant privacy; names and surnames were not recorded, and participants were identified by a numerical code ranging from 1 to 30, with no correlation to their identities. Consequently, the texts obtained from the interviews are anonymous and will not be disseminated. Informed consent was obtained from participants to ensure they fully understand the implications of the research.

2.3. Typology of Interviewees

The thirty interviewees are wine professionals working across four different regions of France: Bordeaux (65%), Burgundy (23%), Loire (6%), and Languedoc-Roussillon (6%). They hold various positions within the wine industry, operating as owners (42%), directors (32%), technicians (13%), and advisors (13%). The study panel is predominantly male (83%) and comprises individuals with diverse educational backgrounds, with agronomists (42%) and oenologists (39%) being the largest groups. While 84% of the panel interviewed practise biodynamics, only 22% are certified (Figure 1).

2.4. Semi-Directive Interviews

During the semi-directive interviews, eleven questions (Supplementary Material, Figure S1) were asked regarding the participants’ definitions of biodynamic agriculture, their practices, and their understanding of the effects of biodynamic preparations on plants, grapes, and wines. The responses were recorded and transcribed in French.

2.5. Textual Analysis

Textual analysis relies on a statistical examination of the words within a text. The development of this tool has been facilitated by the advent of computers. Michel Benzecri is considered the father of textual analysis in France, having been the first to apply multidimensional statistics to textual data [22]. One of his students, Max Reinert, developed a dedicated tool for analysing a set of texts based on a chi-squared matrix [23].

The corpus obtained from the interviews was analysed using Alceste software (Analyse de Lexèmes Co-occurrents dans un Ensemble de Segments de Textes; Reinert, 1986, Image Ltd., Toulouse, France, version 2018). In textual analysis, the corpus is divided by the author into initial context units (UCIs). When preparing the corpus, UCIs are associated with characteristics such as the job position or the educational background of the interviewees. A selection of these characteristics is attached to the relevant UCI and marked with “stars” in the corpus, hence the term “starred words” [24].

A bound morpheme corresponds to a group of text units sharing the same root. For instance, the words “levure,” “levuré,” and “levurage” are replaced by the lexeme “levur.” The reduced terms fall into two categories: analysable terms (nouns, verbs, adjectives, and adverbs) and illustrative terms (prepositions, pronouns, conjunctions, auxiliaries, and starred words) [24].

Next, a hierarchical ascending classification (HAC) is performed using only the analysable terms, dividing the corpus into classes. Alceste employs χ² calculations to measure the probability of a lexeme being used in a particular class. The higher the χ² value between a lexeme and a class, the stronger the contribution of that term to defining the class. Conversely, a lower χ² indicates a weaker association. The terms within each class are presented in descending order of χ² [23,24].

The software does not provide interpretations of the classes; instead, the researcher interprets the classes based on their contents, including the presence or absence of terms (negative χ²), while the starred words contribute equally and solely to the interpretation [24]. A thorough understanding of a class’s content is best achieved by examining the complete lexical field all at once.

The Alceste method is widely used in various contexts; for example, it can be used to analyse collections of scientific reports or evaluations of research topics. Yvette Vaquet and Philippe Jeanne employed Alceste to analyse reports from the International Congress of Arctic Social Sciences (ICASS), assessing continuity and evolution in high-latitude research [25]. Frédéric Brochet analysed expert tasting notes using Alceste in his studies on wine perception, revealing different strategies in expert discourse [26]. Alceste can also be used to analyse patient discourse to understand psychological and emotional functioning. For instance, Villatte and de Léonardis studied the discourse of young adolescents with high intellectual potential to assess their perspectives on the question “Who am I?” They found that gender, age, and socio-cultural background are more significant than intellectual ability in shaping categories of existential reflection [27].

Other studies have analysed patient discourse before and after specific treatments, such as Devienne’s analysis of speech following photo-expression therapy for diabetic patients [28].

All these authors agree that computer analysis using Alceste reveals representations that would otherwise remain concealed. However, these representations may be reductive, as they take into account only semantics, while discourse encompasses other crucial elements—what Kalampalikis refers to as the pragmatic requirements of language. We must not underestimate the influence of relational modes, social conditions, and expressive possibilities in the formation of representations [29].

2.6. Corpus

A single corpus comprising 84,408 characters was created by compiling the thirty interviews in text format. Each of the eleven questions posed to the thirty interviewees was tagged with eight attributes or starred words: job position, educational background, age, gender, region of work, size of property (if applicable), biodynamic practices, and potential biodynamic certification.

Figure 2 provides an example of a response from one interviewee: a young female manager in Bordeaux, practising non-certified biodynamics, qualified in business and viticulture. She is responding to the second question regarding the advantages and disadvantages of biodynamical practices.

3. Results

3.1. Statistical Results of the Text Analysis Using Alceste Software

Table 1. Preliminary results of the corpus analysis using Alceste software.

Analysis Results
Total number of lexemes in the corpus	84,408
Total number of distinct lexemes in the corpus	6385
Number of lexemes used for the analysis	1259
Average number of occurrences of a lexeme	13
Maximum number of occurrences of a lexeme	2727
Number of starred words	77
Number of Initial Context Units (ICU)	339
Number of Elementary Context Units (ECU)	1963
Number of classes obtained by HAC	4

presents the characteristics of the corpus and the initial divisions of the text made by the Alceste software prior to analysis and hierarchical classification. The analysis was conducted on 51% of the text units within the corpus, which is considered a typical score according to the software’s documentation, as the remaining 49% of units are the most infrequent, appearing only a handful of times in the corpus.

Four classes have been defined, using a hierarchical ascending classification (HAC), as illustrated in Figure 3. Each class comprises lexemes that are significantly associated and significantly excluded, with significance determined by the value of χ². A positive chi-square indicates co-occurrence, while a negative chi-square indicates co-exclusion. This allows for the identification of lexemes commonly used within a class and those that are rarely mentioned.

Figure 3 presents the dendrogram obtained by HAC distances between these classes in the form of a dendrogram. In this type of diagram, the groups are connected by branches, and the way these branches join indicates the proximity or distance between the groups. The closer the branches are to the root, the more similar the groups are. In our case, Class 1 appears to stand out from Classes 2, 3, and 4, which form a subgroup. Within this subgroup, Class 2 is distanced from Classes 3 and 4.

3.2. Biodynamic Representations

3.2.1. Biodynamics of Certainty vs. Biodynamics of Questioning

In the French wine industry, two states of mind regarding biodynamics exist. One group (Class 1) expresses certainty about the actions and effects of biodynamics, while another group (Classes 2, 3, and 4) exhibits doubt.

The certainties relate specifically to the effects of biodynamic preparations, particularly silica and horn manure, on plants and soil, respectively. Doubts can stem from various origins and perspectives; for example, some winegrowers (Class 2) question the fundamental principles of biodynamics, leading to scepticism about the potential benefits of this cultivation method. Others (Classes 3 and 4) may lack direct evidence of the effects on the vine or wine yet remain curious and seek convincing facts from practical experience in the field or sensory evaluation in the glass.

3.2.2. Agronomists, Practitioners and Advisors, Convinced of the Material Effects of Biodynamic Preparations

Table 2. The most common and meaningful co-occurrences and co-exclusions of words and starred words defining Class 1.

Co-Occurrences			Co-Exclusions
Words	χ2	Number of times the word appears in the class	Words	χ2
silica	125	78	biodynam	−57
soil	112	101	wine	−54
horn	70	43	organic	−29
plant	69	79	do	−28
effect	65	68	disadvantage	−24
manure	53	34	find	−23
leaves	44	32	was	−21
message	41	23	us	−18
stimulate	35	26	test	−18
sun	34	19	advantage	−16
improve	34	27	agricult	−15
light	29	17	people	−14
structure	28	14	because	−14
photosynthesis	28	15	made	−13
matter	26	22	have	−12
organism	24	20	tank	−12
physics	22	11	practice	−11
growth	22	12	not	−10
action	20	25	yeast	−10
root	20	10	Steiner	−10
Starred words	χ2	Number of times the word appears in the class	Starred words	χ2
*q_actions	143	97	*q_wine-making	−51
*q_effects501	131	83	*q_proscons	−48
*q_effects500	29	68	*q_winerysize_50-100	−27
*winerysize_NA1	24	97	*q_definition	−20
*jobposition_advisor	23	91	*q_wine	−12

NA¹: Not applicable; interviewees do not work in a winery. *: starred-words defining the interviewees characteristics and question type

shows the co-occurrences and co-exclusions in class 1.

The words found in this class are “silica”, “soil”, “horn”, “plant”, “effect”, “manure”, “leaves”, “message”, “stimulate”, “sun”, “promote”, “light”, “structure”, “photosynthesis”, “matter”, “organism”, “physical”, “growth”, “action”, “root system”, “shape”, and “micro”, among others. Most of them are nouns and verbs indicating an action or an increase (Figure 4).

This group makes very little use of the words “Biodynamics”, “Steiner”, “advantage”, “disadvantage”, or the lexical field of winemaking and tasting.

Using the lexemes that define this group, we could write the following sentence: “Silica has an effect on the leaves, bringing a message of light and sunlight that stimulates photosynthesis and plant growth, and structures the plant’s shape. Horn manure acts on the root system and micro-organisms. The effect and action of these preparations are on the physical matter”.

We can therefore assert that a primary representation of biodynamics within the French wine industry is its material aspect: biodynamics has observable and measurable effects. These effects are achieved through biodynamic preparations, which would influence the soil and its microorganisms through horn manure, and affect plant structure via horn silica. This representation appears to diverge from Steiner’s foundational principles, as it does not regard these practices as part of a spiritual philosophy. Such a concrete and materialistic discourse could be applied to any other products, such as foliar fertilisers or crop protection agents.

Table 3. The main characteristics of Class 1 interviewees. The percentages represent the occurrence of each attribute among the interviewees of the class.

Class 1 Composition
Region		Educational Background
Bordeaux	80%	Agronomist	20%
Bourgogne	10%	Agronomist–Oenologist	40%
Loire	0%	Oenologist	10%
Languedoc-Roussillon	10%	Vine and Wine Diploma	10%
Job Position		Practitioners—Certifications
Owner	30%	Practitioner—Biodyvin—Demeter	10%
Director	30%	Practitioner—Biodyvin	10%
Technician	10%	Practitioner—Demeter	0%
Advisor	30%	Practitioner—Not Certified	50%
		Practitioner—NA1	30%
		Not a practitioner—Not Certified	0%
		Not a practitioner—NA1	0%

NA¹: Not applicable; interviewees do not work in a winery.

indicates that in this category, 100% of biodynamic practitioners, whether certified or not, primarily comprise owners, directors, and biodynamic advisors, with a significant proportion being agronomists (60%).

3.2.3. Non-Practitioners, or Uncertified Practitioners, Expressing Scepticism About Fundamentals of Biodynamics

If we apply the same analysis to Class 2, we can identify the most frequently used lexemes: “disadvantages”, “advantages”, “agriculture”, “biodynamics”, “Steiner”, “specifications”, “Agriculture Course”, “definition”, “approach”, “tools”, “risk”, “expensive”, among others. These are common nouns or adjectives that reflect a degree of questioning or judgement, with very few verbs being employed (

Table 4. The most common and meaningful co-occurrences and co-exclusions of words and starred words defining Class 2.

Co-Occurrences			Co-Exclusions
Words	χ2	Number of times the word appears in the class	Words	χ2
inconvenient	109	62	silica	−31
advantage	73	51	effect	−18
agricult	68	39	soil	−17
biodynam	51	148	on	−15
Steiner	47	21	horn	−15
definition	44	20	go	−14
specifications	42	20	small	−13
approach	40	22	an	−12
today	35	31	manure	−12
tools	25	19	leaves	−12
risk	25	18	tank	−10
organic	24	15	is-it	−9
view	23	12	material	−9
point	22	13	year	−8
expensive	19	9	trial	−8
copper	19	11	yeast	−8
question	19	21	pass	−8
course	18	10	message	−8
thing	17	41	stimulate	−8
ecosystem	17	9	improve	−8
Starred words	χ2	Number of times the word appears in the class	Starred words	χ2
*q_definition	106	81	*biodyn_yes	−55
*q_proscons	103	108	*q_actions	−38
*biodyn_no	55	68	*q_effects501	−30
*age_50	22	70	*q_effects500	−20
*certification_NA1	9	91	*winerysize_<20	−11

NA¹: Not applicable; interviewees do not work in a winery. *: starred-words defining the interviewees characteristics and question type

, Figure 5).

In this group, there is a noticeable absence of technical or operational terms related to biodynamics. Words like “silica”, “horn”, “soil”, “effects”, “sun”, “matter”, and “taste” are rarely mentioned.

If we attempt to construct a sentence using the identified lexemes, it could read as follows: “Biodynamics has both advantages and disadvantages. It is an approach linked to organic farming, defined by Steiner, whose ideas are compiled in the Agriculture Courses. Today, numerous questions and perspectives exist. It is an expensive and risky method, and the specifications are restrictive, limiting the use of copper, which heightens the risk of disease.”

This class illustrates a second representation of biodynamics: doubt about its effects and scepticism regarding the anthroposophical foundations of this method. This perspective raises concerns about the relevance of specifications and labels that render these practices restrictive and risky. Doubt and scepticism pose obstacles to the adoption of biodynamics, potentially deterring individuals from trying this production method.

In this group, 90% of the interviewees are either non-practitioners (45%) or non-certified practitioners (45%). The other characteristics follow the overall distribution of the entire panel (

Table 5. The main characteristics of Class 2 interviewees. The percentages represent the occurrence of each attribute among the interviewees of the class.

Class 2 Composition
Region		Educational Background
Bordeaux	64%	Agronomist	5%
Bourgogne	9%	Agronomist–Oenologist	35%
Loire	18%	Oenologist	25%
Languedoc-Roussillon	9%	Vine and Wine Diploma	20%
Job Position		Practitioners and Certifications
Owner	36%	Practitioner—Biodyvin—Demeter	9%
Director	36%	Practitioner—Biodyvin	0%
Technician	18%	Practitioner—Demeter	0%
Advisor	9%	Practitioner—Not Certified	45%
		Practitioner—NA1	0%
		Not a practitioner—Not Certified	36%
		Not a practitioner—NA1	9%

NA¹: Not applicable; interviewees do not work in a winery.

).

3.2.4. Non-Certified Practitioners Are Focused on the Experimental Method to Demonstrate the Effects of Biodynamics on Their Terroir Without Constraints

Class 3 incorporates the following words: “trial”, “research”, “year”, “estate”, “plot”, “service”, “negative”, “people”, “friend”, “discuss”, “organic”, and “biodynamics”. These are primarily common nouns that define specific places and times, along with some dialogue verbs (

Table 6. The most common and meaningful co-occurrences and co-exclusions of words and starred words defining Class 3.

Co-Occurrences			Co-Exclusions
Words	χ2	Number of times the word appears in the class	Words	χ2
test	64	28	silica	−19
year	45	30	plant	−18
organic	44	41	soil	−17
winery	37	19	effect	−11
period	30	12	inconvenient	−11
since	27	14	horn	−10
people	26	26	today	−10
major	25	7	advantage	−8
service	25	8	power	−8
biody	24	8	agricult	−8
winegrower	23	16	enable	−8
plot	22	21	manure	−6
because	22	42	give	−6
discuss	21	6	tools	−6
Burgundy	21	6	living	−6
transition	20	18	matter	−6
negative	20	9	plant	−6
research	20	9	message	−6
back	18	6	use	−6
luck	18	5	organism	−6
Starred words	χ2	Number of times the word appears in the class	Starred words	χ2
*q_practices	89	64	*certification_NA1	−27
*winerysize_50-100	38	54	*biodyn_no	−27
*biodyn_yes	27	213	*q_effects501	−23
*jobposition_owner	20	114	*q_actions	−14
*region_burgundy	15	73	*jobposition_advisor	−13

NA¹: Not applicable; interviewees do not work in a winery. *: starred-words defining the interviewees characteristics and question type

, Figure 6).

There is limited use of terms related to biodynamic preparations and their potential effects, such as “silica”, “plant”, “soil”, “effect”, “living”, and “matter”.

A sentence that could encompass all the lexemes used by this group might be as follows: “To decide to transition from organic to biodynamic viticulture, one needs to conduct trials over several years across various estates and plots. Engaging in discussions with winemakers and other people is essential. This process requires research and time spent with friends to understand the major benefits of biodynamic viticulture, as well as its negative aspects. We are lucky to have this kind of exchange in Burgundy.”

Class 3 presents a representation that contrasts sharply with that defined by Class 1. This perspective lacks certainty, as these winemakers cannot articulate the advantages or disadvantages of specific preparations without a reflective approach tailored to each context. They need to experiment on their own terroir, across different plots and vintages. Acceptance of these methods comes through empirical observation and exchanges with fellow winegrowers and neighbours. No references are made to Steiner or the foundational principles of biodynamics.

In this class, we find that 73% are non-certified biodynamic practitioners, primarily from Bordeaux and Burgundy. They are mostly managers or directors with diverse backgrounds, committed to the experimental method within a specific terroir (

Table 7. The main characteristics of Class 3 interviewees. The percentages represent the occurrence of each attribute among the interviewees of the class.

Class 3 Composition
Region		Educational Background
Bordeaux	64%	Agronomist	18%
Bourgogne	27%	Agronomist–Oenologist	18%
Loire	0%	Oenologist	36%
Languedoc-Roussillon	9%	Vine and Wine Diploma	18%
Job Position		Practitioners and Certifications
Owner	36%	Practitioner— Biodyvin—Demeter	0%
Director	45%	Practitioner—Biodyvin	9%
Technician	9%	Practitioner—Demeter	9%
Advisor	9%	Practitioner—Not Certified	73%
		Practitioner—NA1	9%
		Not a practitioner—Not Certified	0%
		Not a practitioner—NA1	0%

NA¹: Not applicable; interviewees do not work in a winery.

).

3.2.5. Hedonist Practitioners, Whether Certified or Not, Curious About the Effects That Biodynamics Can Have on Wine

Class 4 includes the following words: “grape”, “harvest”, “wine”, “vat”, “indigenous yeast”, “extraction”, “ageing”, “taste”, “savour”, “prefer”, and “smell”. These terms reflect the semantic field of harvest and winemaking, along with verbs related to perception (

Table 8. The most common and meaningful co-occurrences and co-exclusions of words and starred words defining Class 4.

Co-Occurrences			Co-Exclusions
Words	χ2	Number of times the word appears in the class	Words	χ2
wine	130	88	soil	−16
tank	119	30	plant	−9
yeast	95	25	inconvenient	−9
ferment	74	17	life	−7
taste	66	18	part	−7
indigenous	51	9	silica	−7
ageing	50	10	then	−6
pied_de_cuve	46	15	horn	−6
bottle	45	9	advantage	−6
grapes	39	18	allow	−6
harvest	36	14	manure	−5
prefer	34	6	action	−5
smell	29	9	level	−5
wine-making	29	10	organic	−4
drink	28	7	talk	−4
extraction	28	6	simple	−4
vintage	24	12	plant	−4
Brettanomyces	22	4	message	−4
glass	22	7	stimulate	−4
analysis	22	6	vineyard	−4
Starred words	χ2	Number of times the word appears in the class	Starred words	χ2
*q_wine-making	302	73	*q_proscons	−19
*q_wine	79	27	*q_effects500	−18
*certification_demeter	20	37	*q_definition	−16
*jobposition_owner	9	75	*q_actions	−12
*biodyn_yes	7	141	*certification_NA1	−10

NA¹: Not applicable; interviewees do not work in a winery. *: starred-words defining the interviewees characteristics and question type

, Figure 7).

There is also a limited use of terms associated with biodynamic preparations and their potential effects, such as “silica”, “plant”, “soil”, “effect”, “living”, and “matter”.

If we combine all these words into a single sentence, it could read as follows:

“The harvest and wine-making are tailored to optimise the quality of the grapes, with fermentations conducted in tanks using indigenous yeasts through pieds de cuve. The extractions are gentle and followed by tasting and smelling the wine. However, the risk of developing Brettanomyces during ageing or in the bottle can be higher, necessitating more thorough analysis of the wine.”

In this class, we find both certified (18%) and non-certified (82%) biodynamic practitioners, with most coming from Bordeaux. A significant portion are oenologists (63%) holding various positions in the industry, all curious about the effects that biodynamics can have on wine (

Table 9. The main characteristics of Class 4 interviewees. The percentages represent the occurrence of each attribute among the interviewees of the class.

Class 3 Composition
Region		Educational Background
Bordeaux	91%	Agronomist	9%
Bourgogne	9%	Agronomist–Oenologist	27%
Loire	0%	Oenologist	36%
Languedoc-Roussillon	0%	Vine and Wine Diploma	18%
Job Position		Practitioners and Certifications
Owner	36%	Practitioner—Biodyvin—Demeter	9%
Director	36%	Practitioner—Biodyvin	9%
Technician	18%	Practitioner—Demeter	0%
Advisor	9%	Practitioner—Not Certified	73%
		Practitioner—NA1	9%
		Not a practitioner—Not Certified	0%
		Not a practitioner—NA1	0%

NA¹: Not applicable; interviewees do not work in a winery.

).

4. Discussion

Textual analysis using Alceste software has allowed us to describe four representations of biodynamics that currently exist within the French wine industry.

4.1. Semantic Analysis: One Understanding Among Many

Naturally, these representations are not exhaustive. This type of analysis has its limitations. First, it is challenging to assemble a panel that accurately represents such a multifaceted industry. In this case, only 30 individuals were interviewed, primarily from the Bordeaux region. This group of individuals does not encompass all regions, positions, or areas of expertise within the industry. Second, as noted by Kalampalikis and Fracchiolla, discourse is not merely semantic production; it encompasses a complete system of verbal devices, such as redundancy and metaphors, as well as non-verbal devices like body language and silences [18,29]. When viewed as a system, discourse is intrinsically linked to the character and socio-cultural context of each interviewee. Therefore, focusing solely on semantics is reductive [18].

4.2. Biodynamics as a Knowledge Dialogue

The classes identified in this study reveal different representations of biodynamics. Individual opinions and understandings of this method coexist within the French wine industry. According to Jean Foyer, in representations of biodynamics, “we constantly find, but in variously balanced proportions, this tension between different epistemic registers—peasant, scientific and esoteric” [30].

According to Rosset & Martinez-Torres, agroecology is a collective construction built upon knowledge that originated from individual experiences [31]. Boujemaa examined the training and learning of agroecological methods among Bolivian farmers, which was facilitated by scientists, agronomists, and biologists. The learning process involves both assimilation (integrating new ideas into the learner’s trajectory without deconstructing it) and accommodation (transforming the learner’s perspective into a new structure, better suited to incorporate the new ideas presented). Assimilation is most effective when the learner’s prior knowledge is acknowledged. Accommodation, on the other hand, necessitates a confrontation of ideas. Both processes are effective only if the system is willing to move beyond its established norms and accept the need for change. Systems with a strong identity tend to be less open to change through either assimilation or accommodation. Learning becomes richer when it involves structures of prior knowledge that are connected to culture, training, individual trajectories, and the confrontation of different opinions and perspectives [32].

The diverse representations of biodynamics and their coexistence can present an opportunity to foster a new understanding of agriculture during the transition we are currently experiencing. For example, Jean Foyer describes biodynamics as a form of “syncretism,” suggesting that different ways of thinking can lead to new representations and applications [30].

4.3. Biodynamics: Between Traditional and Integral Worldviews

The four classes identified in this study align with some of the agricultural worldviews developed by Cyril Rigolot [17].

The modern worldview corresponds to Class 1, representing a materialist perspective in which science serves as the primary source of knowledge and technology as the means of progress. As Rigolot explains, this rational and objective worldview can create a separation between humanity and nature, which is viewed merely as an instrument [17].

Classes 3 and 4 reflect the postmodern worldview, where contextualised experimentation is the only approach deemed valid. Social exchange among groups with differing visions or perceptions is considered essential for the construction of knowledge. Rigolot associates this worldview with diversity, creativity, authenticity, and imagination [17]. This perspective is often emphasised in the course for farmers, where Steiner referred to the “path of knowledge,” a reflective approach through which each individual constructs their own understanding, tailored to the uniqueness of their context and awareness [1].

Rigolot identifies two additional agricultural worldviews: the traditional and the integral worldview.

The traditional worldview places significant emphasis on religious metaphysics, scriptures, doctrines, and religious leaders [17]. Class 2 of our study critiques the esoteric foundations of biodynamics in their metaphysical aspect. Biodynamics is frequently associated with a religion in the industry, and biodynamic practitioners are often perceived as gurus. This perspective may alienate a considerable number of farmers, as it critiques a biodynamic worldview that closely resembles the traditional one. Alexandre Grandjean, in his study of biodynamic practices in Switzerland, demonstrates that most practitioners distance themselves from the anthroposophical roots of biodynamics [12].

The integral worldview adopts a more holistic approach, integrates systems that are frequently seen as contradictory: science and spirituality, humanity and nature [17]. Unfortunately, this worldview is not represented in our study. What if biodynamics were not a religion, but rather a holistic and transdisciplinary approach that could restore the connection between nature, rhythms, and humanity?

Jean Foyer emphasises two crucial concepts in alternative agricultures such as biodynamics: care and companionship [33]. Practitioners of these methods cultivate bonds with their plants and environment, fostering relationships of reciprocity and affection. They closely observe ecological processes, plant movements, and the development of companion plants, as well as the roles of microorganisms, birds, and bats. They engage with the dynamics of the living world, are trained across a diverse range of disciplines, and remain open to various forms of knowledge. The principles of health, respect, transmission, and sustainability are central to these agricultural practices [33].

4.4. Biodynamic Preparations: The Most Concrete and Practical Application of Biodynamics

Biodynamics is a vast subject, and the scientific community is grappling with how to explore it effectively. One of the aims of this study was to identify the effects of biodynamic preparations based on the empirical perspectives of the sector. This approach could allow research to prioritise analyses and focus on the effects that are most relevant to production.

The findings indicate that, from the empirical viewpoint of winegrowers, the effects of horn silica (501) and horn manure should be assessed in terms of plant growth, canopy structure, photosynthesis, and soil microorganisms. Certain research institutes are beginning to investigate the impacts of biodynamic management, yielding interesting results that support the empirical perspectives of the sector.

Some studies analysing the effects on plants have shown that biodynamically managed vines exhibit shorter shoots, shorter internodes, fewer secondary shoots, and reduced vigour, while maintaining better water use efficiency [34] and lower stomatal conductance [35]. These plants appear to be less susceptible than untreated plants to cryptogamic diseases such as mildew, grey mould, and acid rot [34,35], likely linked to enhanced enzymatic activity on the leaves [36,37].

Regarding soil, some studies indicate that the composition and abundance of microorganisms in biodynamically managed soils are comparable to those in organically managed soils [38]. Others suggest that soils treated with biodynamic preparations host distinct microbial communities, with their microbial composition aligning closely with that of the preparations themselves [39]. These soils demonstrate the presence of growth-promoting microorganisms—those that can enhance plant growth and protect against diseases and abiotic stress by producing hormones [39,40]. Microbial networks appear to differ between biodynamic and organic vineyard management. Ortiz-Alvarez has shown that the microbial networks in soils treated with biodynamic preparations exhibit a high clustering coefficient and few niche specialisations. This characteristic may reinforce soil homeostasis, making it more resilient to the arrival or loss of certain species in response to biotic or abiotic stresses [41].

Current research topics align with the empirical observations of practitioners and could provide valuable insights for those seeking tangible answers regarding this type of viticulture. However, additional field and laboratory studies are needed to assess the effects of this cultural method and to reassure winegrowers who remain sceptical about the principles of biodynamics.

Currently, there are few studies investigating the effects on wine quality.

5. Conclusions

This study has provided an overview of the various representations of biodynamics within the French wine industry. Four perspectives have been described in this analysis, yet many other viewpoints exist within the sector. These diverse perspectives represent a rich resource that can be leveraged and confronted to help agricultural systems evolve.

Steiner’s fundamentals can pose obstacles to the practical application of biodynamics in viticulture. It is crucial to move beyond these foundational principles and develop one’s own reasoning in order to embrace and reinvent these practices. Biodynamics fosters a relationship between the plant, the environment, and the farmer through a transdisciplinary and holistic approach. An increasing number of experimental studies demonstrate the effects of biodynamic preparations on soil microorganisms and plant resilience. To translate these findings into biodynamic practices, collaboration between scientists and practitioners is essential, taking into account the holistic vision of practitioners while defining effective and reproducible methods in the field.

To assist winemakers in developing their own methods based on solid evidence, it is necessary to define clear hypotheses regarding the effects of biodynamic practices by formulating specific, field-testable scenarios and standardising traits and measurement methods to allow for the comparison of datasets over time and across different terroirs. This involves designing controlled experiments; establishing protocols that include both control and biodynamically treated modalities for comparison; collecting quantitative and qualitative data; including biological measurements and practitioner observations; analysing the collected data using appropriate statistical techniques to determine the significance of the results; and publishing and sharing findings for validation and practical application.

A transdisciplinary approach that encompasses all existing representations within the wine industry—such as field experiments, fundamental biological research, participatory studies, and social and anthropological investigations—is essential for the understanding, acceptance, and evolution of biodynamics in viticulture.