Delayed Occurrence of the Result (Dolus Generalis / Weber's Intent)

DELAYED OCCURRENCE OF THE RESULT (DOLUS GENERALIS)^*

Prof. Dr. Murat Volkan Dülger^*

ABSTARCT

The concept of dolus generalis, first introduced into the doctrine of criminal law by an article written by H. B. von Weber in 1825, has since been studied by all scholars concerned with criminal law dogmatics and crime theory. Various solutions have been proposed, and it has also been reflected in judicial decisions. However, it cannot be said that this issue —almost as old as the history of modern criminal law itself— has yet been fully resolved. The reason why this topic, epitomized by the septic tank case of the German Federal Court of Justice (BGH), continues to be intensely debated is that this set of events enables the eclectic discussion of the fundamental components of crime theory, namely objective and subjective imputation, through a single case. Thus, criminal law scholars are able to test both the consistency of their own views and those of other authors through this event. In this article, after first defining the concept as it should be understood, I presented the main opinions and criticisms regarding the solutions proposed to address the resulting problems. I then explained the place of the concept in Turkish criminal law and presented my own views on the matter. Finally, after discussing the so-called “reverse dolus generalis” cases and expressing our perspective on them, I concluded the article by offering my final reflections on crime theory within this context.

Key words: Dolus generalis, Weber’s intent, deviation in causal link, interruption of the causal chain, associating the perpetrator’s intent with the second act, delayed result.

I. Introduction and Presentation Overview of the Topic

Despite numerous plausible theoretical approaches proposed over the past two centuries (1825–2025) to explain the phenomenon known as dolus generalis, a definitive consensus has yet to emerge. Consequently, even after such an extended period of scholarly debate, the issue remains a current and pressing topic within criminal law theory. The significance of this matter, and the persistent attempts to resolve it, lie in the fact that it encapsulates some of the most fundamental and core elements of criminal law dogmatics.¹

In fact, the concept of dolus generalis is no longer used today in its original sense, but rather to designate and define a certain type of event or a cluster of cases.² This term generally refers to situations where the perpetrator believes that the consequence of the offense has been realized through the first act, whereas in reality the consequence occurs through a second act, which is performed merely for the purpose of concealing the initial offense or eliminating evidence thereof. In such cases, although the perpetrator foresees the outcome in a complex sequence of acts, the outcome materializes not through the act he directly intended, but through a different act. Moreover, the causal chain in these cases typically extends significantly longer than in the commission of an ordinary crime. The focus of this discussion is on scenarios where, based on the mistaken assumption that he has already achieved the intended outcome through his initial intentional act, the perpetrator engages in another act without intent, which subsequently leads to the realization of the consequence.³

The classical example of such incidents involves a situation where A acts with the intent to kill B; however, as a result of A’s initial act, B does not die. A, mistakenly assuming that the victim is already dead, attempts to conceal the crime by throwing the body into the water. Subsequent autopsy reveals that the death occurred not due to the initial act, but rather as a result of drowning caused by the perpetrator’s second act.⁴

In this case, whether a completed intentional homicide or merely an attempted homicide followed by a possible negligent killing has occurred constitutes a question that has been the subject of vigorous debate in criminal law doctrine for a long time.⁵ This is one of the oldest issues of criminal law dogmatics,⁶ and although it may initially appear to be an artificial example, there are, in fact, numerous concrete instances of such occurrences in practice. Especially within German and Spanish case law, various forms of this situation are frequently encountered.⁷ In the case law of Türkiye, however, I have not encountered any decisions where this issue has been specifically discussed. Nevertheless, it must be emphasized that the absence of such judicial discussions does not imply that incidents of this nature have not occurred.

In cases of this nature, it must be emphasized that there is no controversy where the perpetrator’s second act is carried out as a precaution, that is, with the purpose of definitively ensuring the result—for instance, where the perpetrator breaks the victim’s neck to eliminate even the slightest possibility that the victim might still be alive.⁸ In such a case, the perpetrator’s intent to kill is clearly present with respect to the second act: the perpetrator shoots the victim with the intent to kill, and although he believes or suspects that the victim might already be dead, he fires another shot to the victim’s head in order to entirely eliminate any possibility of survival, and the victim dies as a result of the second shot. In this situation, a completed intentional homicide is present.⁹ Although there are two physical acts in the natural sense, legally there is only a single act, and it makes no difference whether the result is caused by the first or second act, as they are considered parts of a single course of conduct.¹⁰ However, in the problematic set of cases, the perpetrator believes that the result has been achieved after the first act and does not perform the second act with the intent originally present in the first act. In such cases, there are not one but multiple legal acts, and significant issues arise regarding both objective attribution and subjective attribution between the perpetrator’s initial act and the ultimate result.

The issue of the perpetrator’s subjective attribution —namely, which mental element he is to be held responsible for, which act was committed with which mental element, and which of these contributed to the occurrence of the death— has been extensively debated in the scholarly literature in cases where the perpetrator, acting with intent to kill, performs an initial act that does not result in the victim’s death, but then, mistakenly believing the victim to be dead, carries out a second act intended merely to dispose of the body, with the victim dying as a result of this second act. To illustrate this practically significant debate, let me concretize it through the famous cesspit case (Der Jauchegrubenfall), which was the subject of a landmark decision by the German Federal Court of Justice (Bundesgerichtshof / BGH).

The perpetrator, Ms. F, intending to kill the victim who was involved with her former boyfriend, confined the victim in her home for nearly a full day. During this time, when an argument arose between them, the perpetrator attempted to prevent the victim from screaming by stuffing two handfuls of soil into the victim’s mouth—an act that, according to the first instance court’s findings, was committed with dolus eventualis (reckless intent) to kill. Believing that the victim had lost consciousness and was dead as a result, the perpetrator then attempted to dispose of the “body” by throwing it into a cesspit. A subsequent autopsy revealed that the victim had not died from the initial act of soil being stuffed into the mouth but rather had died from suffocation due to a lack of air after being thrown into the cesspit.¹¹

As can be observed, the main subject of discussion is whether the second act was committed intentionally, or more precisely, whether the intent present in the first act also extended to the second act, and whether the first and second acts should be evaluated jointly or separately. Indeed, it is debatable whether the first act should be classified as an attempted intentional homicide and the second act as negligent homicide, or whether both acts together should be considered as constituting a single completed intentional homicide. Alternatively, it could also be accepted that there are two separate acts, yet the deviation in the causal chain (link)¹² during the second act may be deemed insignificant, and thus the perpetrator could be held responsible solely for the final outcome.

These are clusters of cases that enable brainstorming within the field of criminal law theory, allowing for a comprehensive evaluation of elements such as objective and subjective attribution, causal link, intent and negligence, mistake, attempt, concurrence (joinder of offenses), and deviation, all together. At the same time, they are of utmost practical importance as well.¹³

I would also like to emphasize the following point: The subject I will examine under this heading constitutes a complex area where issues related to both the mental element of the offense and deviations in the causal link are discussed, solutions are sought, and various views are advanced. Therefore, within the systematic structure of criminal law, this topic can be analysed under both the headings of the mental element and mistake. Indeed, in both Turkish and German scholarship, the issue is addressed under the headings of intent as well as deviation in the causal link.

II. Concept and Historical Background

The concept of “Weber’s intent” (dolus Weberisch) originates from Heinrich Benedikt von Weber;¹⁴ however, today it is used in a considerably different sense without establishing a direct connection to its original justification. Therefore, it is first necessary to clarify the confusion surrounding the notion of Weber’s intent. Indeed, Weber’s explanation of dolus generalis and the contemporary usage of the term dolus generalis in Germany, as well as the concept of “Weber’s intent” as used in Türkiye, differ significantly in meaning.

The concept of general intent (dolus generalis),¹⁵ namely the idea that the offender’s intent (Vorsatz) continues to apply with respect to the second act as well, was first elaborated in detail in Weber’s article entitled “On the Different Types of Intent” (Über die verschiedenen Arten des Vorsatzes).¹⁶ This notion was termed dolus generalis by Weber, and through the use of this term, he consciously departed from the prevailing doctrinal distinctions concerning the types of intent that had until then been widely accepted in criminal law theory.

At the time when Weber was developing this view, the prevailing conception of intent held that the perpetrator could only be held criminally liable if they had consciously caused the result (Erfolg) through their conduct (Handlung). According to the legal understanding of that period, it was entirely aliened to qualify as intentional a situation in which the result of death —reflecting the perpetrator’s actual will— occurred through an unconscious act following a deliberate act aimed at killing.

As understood by Weber, general intent (dolus generalis) is clearly distinct from the then-prevailing usage of this term to refer to either “offenses aggravated by result” (Erfolgsqualifikation) or “conditional intent” (Eventualvorsatz). The term “general” (generalis) within this concept does not mean that the perpetrator gave no thought whatsoever to the imputed result, nor does it imply that the result was merely foreseen as one among several possible outcomes. Rather, the concept was employed by Weber to denote “the unification and summation of the individual acts producing the final outcome of the offense, within the totality of a singular criminal volitional tendency.”¹⁷ Weber’s approach was readily accepted at the time, both in legal doctrine (Strafrechtsdogmatik) and judicial case law (Rechtsprechung).¹⁸ This idea of a unifying volitional tendency clearly forms the foundation of the views held by Welzel and Graf zu Dohna,¹⁹ thereby leading to the conclusion of the unity of the act (Einheit der Handlung). However, this theory does not assume the existence of a general intent extending across the entirety of the conduct.²⁰ Rather, Weber’s solution is based on the presence of a “continuing” or “ongoing” intent (fortdauernder Vorsatz) at the stage of concealing the corpse. The acts performed at this stage are also deemed to originate from the perpetrator’s criminal intent (Verbrechensvorsatz).²¹ In Weber’s words, “the perpetrator’s individual volitional tendency represents the synthesis of a single unified course of conduct that constitutes the totality of the offense.”²²

Weber characterizes the view that the result of death cannot be linked to intent as “subtle calculations that run counter to people’s sound understanding” and considers such reasoning as “jeopardizing the prestige of the idea of justice”.²³ In this way, Weber’s proposal gained broad acceptance —at the level of legal scholarship, judicial practice, and even legislation— up until the mid-19th century.²⁴

The classification of the second act (for example, the alleged disposal of the corpse) as an act carried out with intent to kill (Tötungsvorsatz) pushed the boundaries of the concept of intent within criminal law theory and gave rise to growing criticism of this view. As a result, by the mid-19th century, some scholars began to seek alternative approaches to justify the theory of completion. In this context, von Buri²⁵ focused on the causal dimension of the problem, arguing that the eventual death, occurring as the result of the initial intentional act, constituted a causally attributable outcome (verursachter Erfolg) under the “conditio sine qua non” theory (Bedingungstheorie).²⁶ Indeed, this perspective was also adopted in the case law of the time. Mailwald summarized it as follows: “There is an act that was causally relevant for the death, and this act was carried out with intent to kill; therefore, for the German Imperial Court (Reichsgericht – RGH), there is criminal liability for a completed intentional homicide (vollendeter Totschlag).”²⁷

The attempt theory (Versuchslehre), which had been part of the debate from the very beginning and was explicitly rejected by Weber, gained decisive momentum with a major contribution published one year after v. Buri’s article. The author of this contribution, Geyer, sought to demonstrate that dolus generalis was “an unsustainable theoretical construct that neither exists nor ever existed,” and that it should be “discarded as a remnant of criminal law history”.²⁸ Indeed, Mayer described this work as delivering the final blow to Weber’s dolus generalis theory.

In this way, by the early part of the final quarter of the 19th century, the main contours of the debate had been established, and several core arguments had been developed in line with the various proposed solutions.²⁹

The classical version of the completion theory (Vollendungslösung), which I shall examine below (namely the solution developed by Weber), is today grounded in different foundations; however, this issue continues to be addressed under the same name even by those who advocate for the attempt solution, the conformity solution (Übereinstimmungslösung), or even differentiated approaches (differenzierende Lösungen). Thus, the term is used as a general label for such cases and for the various doctrinal solutions proposed in response to them.³⁰ Although the concept is not fully consistent with its current substantive meaning, the debates and positions are still articulated under this terminology.³¹

Today, in cases of this kind, the recognition of a completed intentional offense (vollendetes vorsätzliches Delikt) is no longer justified by reference to the dolus generalis theory, as the notion that there is a single act which —even in its second phase— is carried out with intent to kill has since been rejected. This is because equating a perpetrator’s “desire” or “intention” with the concept of intent, which must be present simultaneously with the act (simultaneitätsprinzip), is fundamentally flawed. Doing so would entail accepting a system of criminal law based on punishing mere thoughts — an impermissible “criminal law of intention” (Willensstrafrecht). Accordingly, the German Federal Court of Justice (Bundesgerichtshof – BGH) rightly holds that the concept of dolus generalis has lost its validity in contemporary law, belongs to legal history, and that extending the intent present at the time of the initial act to later acts is incorrect.³² For this reason, the term dolus generalis —or, as it is more commonly known in Türkiye’s jurisdiction, “Weber’s intent” (Weberischer Vorsatz)— is nowadays used solely for historical and traditional purposes to label such cases, but it no longer plays a role in their legal analysis or justification.³³ Therefore, I have opted to use the term “occurrence of a delayed result” (Verwirklichung eines verspäteten Erfolgs) to better capture the nature of the issue at hand.

III. Questions to Be Addressed in Dolus Generalis Cases

The structure underlying the concept of dolus generalis rests in particular on the assumption advocated by Welzel,³⁴ namely that in such cases there exists, as a whole, a “unified course of conduct” (einheitlicher Geschehensablauf) —a covert killing— and that this process includes intent to kill (Tötungsvorsatz) even at the second stage. However, the view that such cases involve a single, unified act is no longer accepted in contemporary legal thought.

In dolus generalis cases, deviations in the chain of causation (Abweichungen im Kausalverlauf) are particularly contentious. In such scenarios, the perpetrator carries out an act with intent to kill, but instead of killing the victim, the act merely renders them unconscious. Believing the victim to be dead, the perpetrator attempts to dispose of the supposed corpse, and the victim actually dies only as a result of this second act — for example, by being thrown into a river. Accordingly, debates concerning both objective and subjective attribution (objektive und subjektive Zurechnung) revolve around the problem of deviation in the causal link (Kausalitätsabweichung).³⁵

The first point that must be established in this context is that the perpetrator commits a mistake of fact concerning the elements of the offense (Tatbestandsirrtum) in the course of the second act. When the perpetrator throws the victim into a cesspit, they believe the victim to be already dead. In this respect, pursuant to Article 30/1 sentence 1 of the Turkish Penal Code (TCK) (corresponding to § 16/1 sentence 1 of the German Criminal Code – StGB), the perpetrator benefits from their error, as they have made a mistake regarding the factual elements of the offense (objektive Tatbestandsmerkmale). However, under Article 30/1 sentence 2 TCK (§ 16/1 sentence 2 StGB), criminal liability for negligent conduct (fahrlässiges Verhalten) remains applicable. The error in this context originates from the perpetrator’s first act, as the perpetrator assumes that the result of death was already achieved through that initial act. Even if this act was not objectively sufficient or adequate to bring about the intended result (gewollter Erfolg), it must nonetheless be determined whether the final outcome —the death that occurred following the second act— can be causally attributed to the first act.

Second, unlike the majority of cases involving mistake of fact concerning the elements of the offense (Tatbestandsirrtum) in relation to causality, which typically involve only a single act, the present category of cases involves multiple acts. In light of the multi-stage nature of such incidents, it becomes necessary to examine whether the second act —producing the final result— can be objectively and subjectively attributed to the perpetrator in light of the first act (from a dogmatic perspective), and whether the perpetrator can be held liable for a completed intentional homicide (vollendeter vorsätzlicher Totschlag) (from a teleological perspective, particularly in terms of the sense of justice).³⁶

The questions to be answered in light of this type of case are as follows:

a) Whether there exists a completed intentional offense (vollendetes vorsätzliches Delikt),

b) Whether the initial act should be classified solely as an attempted intentional killing (versuchter Totschlag), and consequently, whether the second act constitutes negligent causation of death (fahrlässige Tötung),

c) If the first option is accepted, how its theoretical foundation (i.e., its consistency with criminal law theory and justification) can be established.

In order to answer the general questions above and to determine criminal responsibility, the following fundamental questions must first be addressed:

a) Did the risk created by the perpetrator, in line with their initial intent (Anfangsvorsatz), ultimately materialize?

b) Does the deviation in the causal link (Kausalverlauf) pertain to the originally intended risk or to an entirely different risk?

c) Did the perpetrator’s act create a new risk that is independent of the initial risk?

Answering these questions reveals the perspective of doctrinal opinions, particularly in the interpretation of deviations in the causal link. For example, Roxin’s approach, when assessing whether deviations align with the perpetrator’s plan, focuses not on the perpetrator’s intent but on the objectively realized risks.

This demonstrates that the perpetrator’s liability is connected not only to intent or will but also to the consequences of the concrete risk created.³⁷

IV. Opinions and Critiques Developed by Doctrine and Jurisprudence

There are numerous theories that attempt to explain cases of dolus generalis. Almost every scholar who has written on this subject has sought to resolve the issue according to the structure of their own theory of crime. However, these views can essentially be divided into three main categories: the completion theory, the attempt theory, and the adequacy theory.

In response to the justificatory deficiencies inherent in the adequacy theory, a considerable number of alternative approaches have been developed within the doctrine. These are collectively referred to as “differentiated theories.” Each of these theories will be set out below, accompanied by the main criticisms levelled against them. Furthermore, particular attention is given to Roxin’s theory of the perpetrator’s realization of their criminal plan (Planverwirklichung), which, although classified among the differentiated theories, warrants separate and detailed examination.

A. Completion Theory

According to Weber’s original, or in other words, classical doctrine of dolus generalis, it is defined as “the aggregation of all acts constituting the offense and the totality of the volitional orientation indicative of a criminal intent.”³⁸ In line with this, Welzel assumes that the second act also falls within the scope of the intent to kill.³⁹ Accordingly, the perpetrator performed the initial act with the intent to kill the victim. The reason for the second act is causally linked to the first; had the perpetrator not carried out the first act, the second act would not have followed. Moreover, if the perpetrator had realized that the victim did not die as a result of the first act, they would have repeated the initial act or performed a second act to achieve the fatal result. Thus, the perpetrator initiated the course of conduct and ultimately attained the intended result not through the initial intentional act but through the negligent second act. In this case, the theory of the “merging of the causal link” should be accepted, as there is merely an insignificant deviation in the causal chain.⁴⁰ This deviation should not be taken into account, and the perpetrator should be held responsible for the fatal result initially intended and be convicted of completed intentional homicide.⁴¹ Although this theory harbours theoretical inconsistencies, it provides a solution that satisfies the sense of justice in terms of its outcome.

The completion theory is subdivided into several sub-theories: the act-based solution, the causality-based solution, and the adequacy-based solution.⁴² The first two will be examined under this heading, while the latter — which has become predominant in doctrine and practice — will be analysed under a separate heading.

1. Act-Based Solution

a. Explanation of the Theory

Within the act-based solution (Handlungslösung), the entire sequence of conduct —from the attempted killing of the victim to the concealment of the “corpse”— is viewed as a single act, and thus the existence of a single criminal intent (dolus) encompassing this act is acknowledged. Accordingly, in cases of dolus generalis, it is accepted that there is only one continuous act and, consequently, only one offense. This can be understood either in the sense that the perpetrator’s intent extends over both individual acts, as suggested by Weber, or — even if the second act is considered outside the perpetrator’s original intent— that a single intentional killing is recognized due to the connection between the acts.⁴³

According to Puppe, only the initial act was carried out with the intent to kill; therefore, only this act can form the basis for the punishability of the offense, as it is impossible to harbour an intent to kill a person presumed to be already dead. This initial act alone constitutes a sufficient basis for punishability. It not only constitutes the execution of a single act but also a completed attempt. All the facts that establish the completed attempt —and thus the danger inherent in the intent— are fully and continuously present in the causal explanation of the result. The perpetrator threw the victim into the cesspit because he believed the victim was already dead; he was able to do so because the victim was no longer capable of defending himself. The victim’s helpless state was due to the perpetrator’s use of a generally effective killing method, namely suffocation with sand. The perpetrator is not required to foresee all factors in the causal process in advance; hence, even if there are factors within the causal chain that the perpetrator did not consider causally significant, the existence of such factors (in this case, drowning in the cesspit) is irrelevant and does not preclude holding the perpetrator responsible for the result. The fact that these factors were not committed with intent does not convert the perpetrator’s previously intentional conduct into mere negligence. The intentionally executed and result-producing acts by the perpetrator are to be regarded as natural causes.⁴⁴ Therefore, Puppe holds the view that the intent associated with the initial act also encompasses the result that ensued from the second act committed negligently. This conclusion is consistent from Puppe’s perspective, as she considers intent to consist merely of dolus eventualis (conditional intent). Accordingly, all outcomes accepted by the perpetrator (as in this case) must be regarded as falling within the scope of intent.

b. Critique of the Theory

The act-based solution appears insufficient in terms of theoretical justification. If what is meant by the perpetrator’s intent is that the perpetrator’s dolus was still present during the concealment of the corpse, this solution is refuted by the arguments of the attempt-based solution, which will be discussed later. For if the perpetrator genuinely believes that the victim is already dead, it is illogical to assert that he still intends to kill him. It is irrational for the perpetrator to intend to kill someone he already believes to be dead. If what is meant by the perpetrator’s intent is his general mindset, then this is not a proper solution either. Indeed, the perpetrator’s second act undeniably caused the victim’s death; had the perpetrator known that the victim was still alive, he likely would have performed the second act — perhaps even with greater determination. Consequently, this solution emphasizes not the perpetrator’s intent regarding the act itself but rather his overall mental state concerning the entire event. Therefore, it is deemed irrelevant whether the fatal outcome resulted from the perpetrator’s initial act, which remained at the stage of attempt, or from the second act, likely carried out with negligence. This is further evidenced in the proposals of scholars who argue that if the perpetrator exhibits a negative attitude toward his initial act —that is, if he undergoes a change of mindset prior to the second act— the offense of completed homicide cannot be considered as having occurred.⁴⁵

Maiwald, offering a sharp and profound critique of this view, argues that such reasoning could only be correct if the perpetrator’s mental state were considered a relevant factor in assessing the entirety of conduct comprising multiple criminal offenses. For example, a series of acts —kicking, slapping, and punching a person— is regarded as a “series of beatings” (Prügelserie), and collectively constitutes a single offense of intentional bodily harm. Although different acts of intentional bodily harm are committed, they are perceived as manifestations or externalizations of a single criminal intent. However, it becomes problematic to speak of such unity where, although the perpetrator might have accepted or even desired certain outcomes while performing specific acts, he was unaware that these outcomes were actually caused by his conduct. Knowledge that one is infringing upon another’s legally protected interest is a crucial element that unites the externalized orientation of criminal intent. For instance, in a case of intentional bodily harm, if the perpetrator assaults the victim several times and then drives away in his car, unintentionally hitting the victim without even contemplating such a possibility, one cannot speak of a unified intent covering the vehicular impact. Since the perpetrator did not actually foresee this outcome, his current (actual) intent does not amount to a conscious decision to violate the legal norm (an updated malicious intent) and, therefore, does not constitute a criminal intent of the same nature as that underlying the previous intentional acts.⁴⁶

In his critique of this view, Schroeder presents an example in which he generally assumes a completed intentional killing in dolus generalis scenarios, yet arrives at the following conclusion: if the perpetrator, believing the victim to be dead, loads the body onto a truck and, while attempting to bury it in a remote location, the victim dies in a traffic accident en route, a “significant deviation” from the anticipated causal process occurs. In such a case, a genuine concurrence (real concurrence — reale Konkurrenz) arises between attempted intentional killing and negligent manslaughter. This is because an act committed through unconscious negligence cannot form a unity of action (Handlungseinheit) with the commission of an intentional crime; an offense can only be committed either intentionally or negligently, not both.⁴⁷ Extending this example, the question is posed: if no traffic accident occurs and the victim is instead transported to the planned location and buried there, causing death as intended, should a unity of action be accepted between the attempted killing and the burial? The answer given is that if the second act (e.g., a traffic accident) occurs, it must be assessed in real concurrence with the initial act. This example also clearly demonstrates that the temporal proximity between the two stages in dolus generalis cases has no impact on the legal evaluation. Indeed, whether the victim is thrown into the water immediately after the attack or the body is concealed the following day through an act of hiding makes no difference in the legal outcome.⁴⁸

Roxin has also criticized this perspective. According to him, while these traditional views contain a valid core, they lack sufficient conceptual clarity. It is evident that the unity of the offense —encompassing the criminal tendency and the intent to kill— is completed with the initial act, and that the second phase of action falls outside this scope. Treating the concept of dolus generalis as a theory that merely connects various acts exposes the shortcomings of approaches that attempt to argue that the second phase still involves “will” or “intent.” In this context, attempts to reconstruct psychological elements that are no longer present do not lead to correct conclusions.⁴⁹

Roxin has also criticized the views of scholars who approach the doctrine of dolus generalis from the perspective of the attempt theory. In this context, he asserts that while the solution based on causal deviation correctly grasps the dogmatic and logical structure of such cases, this perspective errs by detaching the problem from its teleological context and focusing solely on the deviation in causality. In particular, it is misleading to assess causal deviations merely from a subjective standpoint, neglecting the objective framework of the perpetrator’s plan. Such an approach results in outcomes that are inconsistent with the perpetrator’s objective will and plan.⁵⁰

Roxin has also scrutinized —or more accurately, put to the test— the prevailing view in doctrine and practice concerning the scope of intent. According to him, the question of whether the perpetrator’s success in achieving the result through the second act can be attributed to his intent (subjective attribution) —and thus whether a completed offense can be recognized— essentially depends on whether the perpetrator possessed an intent to kill during the initial act or acted with dolus eventualis (conditional intent) with respect to the fatal result. A result is deemed to have been brought about intentionally only if it was realized according to the perpetrator’s plan. From this perspective, if the initially intended result occurs following an inadequate causal link, the result can no longer be attributed to the perpetrator’s intent. In such cases, the result is the product of coincidence and not a realization of the perpetrator’s design. Accordingly, for example, if the perpetrator, believing the victim to be dead, loads him into a car intending to bury him in a remote location, but the victim dies during the journey as a result of a traffic accident, at most an attempted intentional homicide can be considered.⁵¹

In my view, before evaluating the act-based theory, it is first necessary to clarify what is meant by the unity of an act. In terms of the unity or plurality of acts, a twofold distinction is made: (i) the concept of a natural act and (ii) the concept of an act in the legal sense. A natural act constitutes the basic form of all acts. For example, a slap delivered by A to B constitutes a single natural act. However, in certain cases, even if there are multiple natural acts, there may still be a single act in the legal sense. This situation, referred to as the unity of acts in the legal sense, is itself subdivided into two categories: typical unity of acts and natural unity of acts. Typical unity of acts arises when the legislature requires the commission of multiple acts for the offense to be constituted. A paradigmatic example is the composite offense of robbery, which requires the simultaneous commission of both intentional bodily harm and theft. In contrast, natural unity of acts occurs when multiple acts directed toward a single purpose are performed under a single volitional decision and within the scope of a single criminal intent. For instance, repeatedly stabbing a person is regarded as a single act under the principle of natural unity of acts.⁵²

Cases of dolus generalis can be explained using the classical example provided above. If the perpetrator shoots the victim with the intent to kill and, believing the victim to be dead, proceeds to bury him, two natural acts are formed. However, it is also necessary to evaluate whether there is a single act in the legal sense. Examining typical unity of acts, it is clear that for the offense of intentional homicide, the legislator does not require the commission of multiple acts. Therefore, typical unity of acts is not present. The final and most relevant issue in this context is the natural unity of acts. At first glance, it could be argued that the perpetrator acted under a single volitional decision and within the scope of a single intent aimed at a singular purpose, thus constituting a single act. However, if the perpetrator’s intent changes and a new criminal activity commences, a new and independent act arises, which precludes the qualification of the conduct as a single act.

2. Causality-Based Solution

a. Explanation of the Theory

The second theory under this heading is the “causality-based solution.” The foremost proponent of this theory is von Buri, who argues that there is a single intentional killing in such cases. He asserts that the causality of the intentionally committed initial act, in terms of the theory of conditions (condicio sine qua non theory), leads to the ultimate fatal result. Therefore, the perpetrator should be held liable for intentional homicide.⁵³

According to this view, the resolution of the issue lies in the causal link between the perpetrator’s initial act and the occurrence of the result. This causal connection is grounded in the perpetrator’s initial intentional conduct. This perspective is particularly reflected in the reasoning of a decision by the German Imperial Court (Reichsgericht, RGH).⁵⁴ Based on the established jurisprudence of the RGH’s Criminal Divisions, the judgment proceeds from the principle that “a cause of a result is any act without which the result would not have occurred,” thereby acknowledging that the fatal result is causally linked to the perpetrator’s act of assault. This view is explained in the judgment with the following statement:

“Because the cause of the drowning (death by submersion) was the situation into which the defendants had placed the woman (i.e., the victim); they submerged her because they believed she was already dead and intended to conceal the crime committed against her. However, the will that guided the defendants’ prior assault was directed toward killing the woman. Whether her subsequent drowning was the direct cause of her death is irrelevant, for in any case, the death occurred as a result of the defendants’ will.”

Accordingly, the act causally linked to the death was carried out with the perpetrator’s intent. Therefore, according to the RGH, there is liability for an intentional killing.⁵⁵

b. Critique of the Theory

Proponents of the causality-based theory argue that, according to the theory of conditions, the result would not have occurred without the perpetrator’s initial act, thereby establishing a causal link between the act and the result. According to the theory of conditions, a result occurs through the conjunction of all necessary conditions that must have existed beforehand. Each of these conditions is necessary for the occurrence of the result and holds equal value with the others. However, the understanding that “had his act not occurred, the result would not have ensued” risks rendering almost everyone liable for the acts of others. Thus, the theory of conditions, which recognizes even remote and insignificant conditions as causal, has been restricted by the doctrine of objective attribution. I find the reasoning that the perpetrator can be held liable for a completed offense merely by establishing a causal link under the theory of conditions to be insufficient and incompatible with the principles of modern criminal law.

The assumption advanced by the causality-based solution —that causality between the perpetrator’s act and the result, combined with the presence of intent at any point during the act, suffices to constitute intentional homicide— does not operate so simplistically. The mere fact that the perpetrator had, at some point, the will to kill within the causal chain leading to the fatal result is insufficient. Were this the case, a person who, intending to commit murder, purchases a firearm and, the day before the planned murder, accidentally discharges the weapon in a crowd —unknowingly killing the intended victim— would be liable for intentional homicide. The flaw in this view lies in its failure to recognize that, under crime theory and criminal law, only intent contemporaneous with the commission of the offense can generate legal responsibility. Just as the cognitive element of intent requires that the offender be aware of the elements of the offense at the time of its commission (Turkish Penal Code Art. 21, 30/1; German Criminal Code § 16/1), the volitional element of intent is legally significant only if it exists at the moment the criminal act is committed. The conclusion that follows is that in dolus generalis cases, a perpetrator’s act can only be deemed intentional homicide if, in accordance with the principle of simultaneity, the perpetrator wills the fatal result while carrying out the acts constituting the offense of intentional killing.⁵⁶

Regardless of which completion theory is adopted, it is undisputed that the initial act —the assault against the victim— constitutes an act of execution of the offense of homicide. However, these theories leave unanswered the question of how and why a subsequent act, carried out without intent, can combine with the earlier act, which remained at the attempt stage, to constitute the offense of intentional killing.⁵⁷

Otto, who completely rejects the theory of intent —also known as dolus generalis— argues that there is no need to resort to the dolus generalis doctrine, which excessively broadens the perpetrator’s liability in such cases. He asserts that, in this situation, the perpetrator, under the assumption that the victim was still alive, committed an act involving subjective attribution by throwing the victim into the water, thereby completing a coherent and intentional act. Thus, in such circumstances, there is no necessity to rely on the dolus generalis doctrine, which leads to erroneous conclusions.⁵⁸

B. Attempt (Separation) Theory

a. Explanation of the Theory

According to the attempt or separation theory (Versuchslösung / Trennungstheorie), since the causal link between the act and the result is separated, the case must be analysed in two distinct parts.⁵⁹ In this context, the act should be assessed separately in relation to the result intended by the perpetrator. With respect to the first act, since there is no fatal result attributable to the perpetrator, but he acted with the intent to kill, he should be held liable for attempted intentional homicide. Regarding the second act, as the perpetrator mistakenly believed that the subject of his conduct was a corpse, he should be held liable for negligent homicide based on an error concerning the elements of the offense.⁶⁰

The perpetrator, through the initial act, acted with the intent to kill the victim; however, despite committing the act, the intended result did not occur due to circumstances beyond his control. Therefore, with respect to the initial act, the perpetrator is liable for attempted intentional homicide. With the second act, the perpetrator intended to dispose of the victim’s body, believing the victim to be dead. There is both a causal link and objective attribution between the perpetrator’s second act and the resulting death. However, the perpetrator did not perform the second act with the intent to kill; rather, his intent was directed toward disposing of the corpse. Nonetheless, the perpetrator was expected to exercise objective care concerning the possibility that the victim might still be alive. His failure to exercise such care resulted in the victim’s death through negligence. According to this view, even though the perpetrator was under a mistake regarding the material elements of the offense (Article 30/1), since the second act was negligent and the mistake merely negated intent, the perpetrator must be held liable for negligent homicide. Consequently, the perpetrator should be held liable for attempted intentional homicide based on the first act and for negligent homicide based on the second act.⁶¹

This view is based on a pure dogmatic understanding. Its main argument is that there must be congruence —that is, simultaneity— between the objective and subjective elements of the offense.⁶² Accordingly, the fatal result can be objectively attributed only to the final act; however, in the final act, there is no intent to kill, because no one intends to kill a person they believe to be already dead. Therefore, in such cases, completed intentional homicide cannot be established. Although the initial act was committed with the intent to kill, it did not cause the fatal result and is thus classified merely as an attempted homicide.⁶³

In general, proponents of the attempt-based solution defend their position by rejecting the previously mentioned arguments.⁶⁴ They particularly deny the objective attribution of the death to the initial act, arguing that the result was not the realization of the risk created by the initial act, but rather the realization of a different risk. Due to the risk created by the second act, the “risk connection” necessary for objective attribution in relation to the initial act is severed.⁶⁵ In other words, there is a deviation in the causal process, but it is a significant deviation, and the conclusions drawn may lead to unjust outcomes. Furthermore, proponents of the attempt theory argue that characterizing the killing as completed intentional homicide merely because the perpetrator initially had an intent to kill and the victim subsequently died reflects a moralistic approach that relies solely on the initial intent without further assessment of the concrete sequence of events.⁶⁶ They emphasize that such an approach, which bases liability on the perpetrator rather than the act, constitutes a regression from the act-based focus of modern criminal law. From this perspective, the notion of punishing the perpetrator based on his initial intent is criticized as a reversion to a perpetrator-based criminal law, posing a danger despite its superficial appeal to the sense of justice.

In this case, it is clear that the perpetrator did not possess an intent to kill with regard to the second act, as it is not plausible to claim that the perpetrator intended to kill a corpse. Thus, it cannot be accepted that the perpetrator envisaged the elements of the offense at the time the act causing death was performed. Consequently, the requirement under the principle of simultaneity —that intent must exist at the moment the offense is committed— is not fulfilled.⁶⁷ Moreover, against the argument that the event as a whole constitutes a single act, it is counter-argued: “However, the assumption of unity of action can be refuted on the grounds that the successive acts stemmed from different volitional decisions. Furthermore, it cannot be determined whether the perpetrator would have been willing to continue acting had he correctly perceived the failure of the initial act. In cases where the perpetrator does not accurately perceive the result and the result does not occur through the initial act, it is impossible to attribute the result to the perpetrator’s intent. In such a situation, the perpetrator’s liability would most likely rest on an attempted intentional offense combined with a completed negligent result offense.”⁶⁸

This solution is based on the principle of simultaneity, which requires that intent and action coincide. Indeed, proponents of this view agree with the criticism that, in the cesspit case, the Bundesgerichtshof’s (BGH) dolus generalis doctrine does not align with the principle of simultaneity, as the perpetrator lacked intent during the second act.⁶⁹ Accordingly, liability arising from the intent present prior to the act (dolus antecedens)⁷⁰ should be rejected, as this view opposes a covert intention-based criminal law (Gesinnungsstrafrecht), which relies on the “versari in re illicita” doctrine—holding that one who engages in unlawful conduct bears responsibility for any resulting consequences. According to proponents of the attempt theory, the effort of the completion theory to unify the two acts into a single act and to assert the existence of a unified offense is neither correct nor necessary.⁷¹

Ozansü also asserts that the matter should be resolved within the framework of attribution (Zurechnung), emphasizing that it is necessary to address whether the perpetrator was aware of various possible versions of the causal chain at the time of acting. According to him, attribution is possible only when the perpetrator is conscious of the course of action, meaning there is a fundamental qualitative difference between acting under the suspicion that the victim is still alive and acting under the belief that the victim is already dead. In the first case, the perpetrator still generates a causal chain through his own action, whereas in the second case, the causal chain is influenced by an external natural factor independent of the perpetrator’s will. Just as one cannot form intent toward an already accomplished result or toward an outcome entirely beyond the realm of life experience —where at most a mere wish can exist— similarly, a perpetrator cannot be held liable under the doctrine of intent for the result produced by an action of which he is unaware. The author, thus, reaches a conclusion similar to that of the separation theory (Trennungstheorie), stating that the first act remains within the domain of attempt.⁷²

Maiwald notes that the core argument advanced by the proponents of the attempt solution —namely, that the “actual act of killing” is carried out without intent— is no longer disputed by the opposing view, and its theoretical correctness is now generally accepted.⁷³ According to this doctrine, which has been revisited and further elaborated particularly by Maiwald,⁷⁴ in all such cases, only an attempted killing (attempted murder) can be said to exist; therefore, there can only be a real concurrence (reale Idealkonkurrenz) between the attempted killing and the possible negligent homicide committed through the second act.⁷⁵

This view has also been defended by Maurach: “If the perpetrator, when carrying out the second act, proceeds under the definite assumption that the offense has already been completed, then conceptually no intent can be assumed regarding the second act, for a dead person cannot be killed. In such a case, the perpetrator is attributed with an already concluded intent.” The argument advanced in the doctrine—the general process adequacy argument, a differentiated version of the completion theory—should not overlook the fact that the decisive condition arose after the intent had already ceased.⁷⁶

Similarly, Schroeder⁷⁷ emphasizes that this is not a matter of a deviation from the perpetrator’s conceptualization of the causal process, but rather an issue concerning the absence of intent during the perpetrator’s commission of the typical act of the offense.⁷⁸

Another justification for the attempt theory was developed by Sancinetti:⁷⁹ If the perpetrators of a murder bury, drown, burn, or hang a victim presumed to be dead in a manner causing death, it does not constitute the realization of the initial risk but rather the realization of a different risk; therefore, due to the non-realization of the initial risk, only negligent homicide is recognized. The argument that the structure of the two risks is identical is not convincing. According to Sancinetti, if a completed offense is to be discussed, it must be foreseeable from the outset that death could occur through the second act. However, if the unlawful risk of the second act was already inherent in the first act, it remains unexplained why the resulting consequence is not regarded as the realization of the initial risk but instead as the realization of an entirely different risk. If an attempt is accepted in cases of dolus eventualis (reckless intent), this situation is not due to the non-realization of an unlawful risk already present in the first act, but rather the absence of the realization of a plan, because in order for a result to be attributed to a specific offense element, the execution of a plan must additionally be realized.⁸⁰

In cases of dolus generalis, proponents of the attempt theory are compelled to separately define the second act and determine its relationship with the first act. Since the result of death cannot be (intentionally) attributed to the first act, it can only be classified as an attempted murder. However, this does not automatically mean that the death occurred as a result of the second act through negligence, because this depends on whether the characteristic elements of a typical realization are present. This point was previously noted by Geyer,⁸¹ but as emphasized by Maiwald, it was Pfotenhauer, as early as 1839, who correctly formulated the second phase, which lies at the core of this problem: “The command to the perpetrator acting with the intention to kill should be stated as follows: If you intend to kill someone and believe you have killed, before proceeding to a second act for the purpose of concealment, you must ascertain with certainty whether death has actually occurred through this act, which by its nature is suitable for killing; because only with this second act can you achieve your aim.”⁸² Therefore, if it is accepted that cases of dolus generalis result in negligent homicide, the relationship between the second act and the first act must be determined. The vast majority of those who advocate for this solution⁸³ accept that there is a situation of real concurrence of offenses (cumulative sentencing) here.⁸⁴

b. Critique of the Theory

Roxin states that the view which advocates that, in cases of dolus generalis, only attempted intentional homicide and possibly a negligent offense should be accepted in every case must be rejected. This view particularly relies on the assumption that the intent (mens rea) must exist “at the moment of the act” (principle of simultaneity) and therefore asserts that intent is lacking in the actual killing act (the second act). Consequently, according to this view, the perpetrator is being held responsible based on an intent that had already ceased. However, according to Roxin, this perspective is incorrect because it is necessary and sufficient for the intent to exist not throughout the entire process, but merely at the moment when the perpetrator loses control over the causal course of events. In cases of dolus generalis, the victim’s death should be regarded as a consequence corresponding to the perpetrator’s act committed with the intent to kill, provided that the result can be seen as the realization of the perpetrator’s plan, and this is sufficient for it to be associated with intent.⁸⁵

Puppe states that, as a result of the decision rendered by the German Federal Supreme Court in the cesspit case mentioned above, the opinion put forward in the doctrine —that in such cases only attempted intentional homicide and negligent homicide occur, and that a complete offense cannot be constituted by combining the attempt with negligent homicide (the attempt theory)— has been refuted.

The attempt theory is theoretically accurate and consistent with criminal law dogmatics, but it does not align with the sense of justice. It raises questions such as: “Is it fair to mitigate the penalty merely because the perpetrator ultimately caused a death, yet it is classified as an incomplete attempt? Should the legal order be based on a moral understanding that glorifies chance?”⁸⁶ and leaves these questions unanswered. Indeed, even some proponents of the attempt theory perceive it as conflicting with the sense of justice. Among these scholars, Gropp concedes that “chance (whether good or bad) cannot be considered to the detriment of the perpetrator, but must be considered in their favour,”⁸⁷ and states: “In dolus generalis cases, the perpetrator is ‘lucky’ because success is not achieved at the outset. As with every attempt, this failure should be assessed in favour of the perpetrator.”⁸⁸

Although it is true that there is always a “chance” in (completed) attempts from the perspective of criminal law, it is not correct to consider this solely as an advantage, nor can it be said that this chance is justified. It should not be regarded as an advantage because the same coincidental circumstance that leads to the failure of person A’s attempt may cause the completion of a similar (parallel) attempt by person B, resulting in B being punished due to this misfortune.⁸⁹ The task of criminal law, however, is to distinguish between injustice and misfortune.

Although the attempt theory appears consistent from a theoretical standpoint, it nonetheless harbours significant inconsistencies. There is no debate regarding the fact that the perpetrator committed the initial act with intent to kill but failed to achieve the result due to circumstances beyond their control; all opinions agree on this characterization of the first stage. The attempt theory asserts that the perpetrator acted negligently during the second act and that the resulting death occurred due to this negligence. In our view, the problematic —or at least highly questionable— point lies in whether the perpetrator’s second act was truly committed negligently. In other words, can the death resulting from the second act be subjectively attributed to the perpetrator based on negligent behaviour? Step by step, we see that the perpetrator’s initial act was carried out with direct or eventual intent. But if a perpetrator intends to kill the victim, uses appropriate means, presumes the result has been achieved, and then —based on a mistaken belief— carries out a second act, can it be reasonably expected for the perpetrator to check whether the victim is actually dead and act accordingly? Since negligence involves a breach of an objective duty of care, for it to be said that the perpetrator acted negligently in the second act, it must be established that there was a violation of such a duty. But is it appropriate and reasonable to impose this duty of care as an element of negligence on the perpetrator in this context? In my view, it is not. Expecting a perpetrator to check the outcome of an act committed with intent to kill and to exercise such care presents a major conflict with criminal policy. Moreover, if such care were expected and exercised, a perpetrator who realized that their first act had failed to cause the intended death could intentionally carry out a second act to ensure the result. If such care is not required, the victim might still have a chance to survive. From a criminal policy perspective, the paramount value is the protection of human life. If such a duty of care were imposed, this value could not be adequately safeguarded. Therefore, when no such duty of care is required and this perspective is applied to the case, it is not possible to subjectively attribute the result of the second act to the perpetrator. Hence, the attempt theory contains a significant internal contradiction. Although it may appear theoretically consistent, it ultimately conflicts with the principles of criminal policy.

A strong counterargument to my view could be that if an obligation were imposed on the perpetrator to check whether the victim was alive, upon discovering the victim was indeed alive, the perpetrator might desist from completing the act and engage in voluntary renunciation, thus taking actions necessary to save the victim’s life. Consequently, it could be argued that my criticism involves an implicit act of “mind reading” of the perpetrator’s intentions. This critique is partially valid; however, it is important to note that the same mind-reading issue applies to the counterargument. Assuming that the perpetrator would abandon the act instead of completing it is just as much a form of mind-reading as my position, making it difficult to treat this as a valid counterargument. In my view, even if the perpetrator acted with eventual intent, if they moved to kill a person and completed their attempt, the probability that they would proceed to finish what they started upon discovering the victim was alive is greater than the likelihood of voluntary renunciation. Ultimately, it comes down to a probability assessment: which would save more lives — the scenario where the perpetrator is expected to check on the victim’s condition and possibly renounce, or the scenario where the victim is left to their fate without such a duty? It is impossible to answer this question with certainty. Therefore, adopting a teleological perspective and considering the protected legal interest theory, it is preferable that no duty of care be imposed on the perpetrator. If this perspective is accepted, there would be no duty of care, and consequently, it would be inappropriate to speak of negligent behaviour on the part of the perpetrator regarding the second act.

In our view, another aspect of the attempt theory that deserves criticism is that by completely separating the acts from each other, it overlooks the natural connection between the two acts and fails to take into account the consequences within this context. Although I believe that a modern criminal law system, which prioritizes the act over the perpetrator, should be valid, ignoring the fact that both acts in dolus generalis cases are carried out by the same perpetrator entails serious dangers such as the impossibility of punishing the perpetrator. Even though the perpetrator may not have achieved the intended result with the first act, I think that in the vast majority of dolus generalis cases, the first act significantly contributes to the realization of the result. Perhaps if the first act had not been carried out by the perpetrator, the second act alone would not have been sufficient to achieve the result. Therefore, even if an attempt is made to completely separate the two acts, it is not possible. Moreover, by ignoring the fact that a completed crime is constituted by the totality of both acts, there arises a significant danger: the perpetrator may receive a punishment much less than what the realized result requires, leading to an unjust outcome.

At this point, it should also be noted that there may be cases of dolus generalis in which the application of the attempt theory would be appropriate. However, for this to be the case, there must be a temporal and spatial difference between the first and second acts. Thus, the natural connection between the two acts would lose its significance, and the attempt solution could be a fair application. Specifically, if a relatively long period of time passes after the first act and if the second act takes place in a different location from the first, causal processes of a nature that cannot be deemed insignificant may intervene. Furthermore, due to the time elapsed between the two acts, the assumption that the perpetrator initially acted with the intention of killing the victim and would in any case proceed to kill becomes weaker. However, it must be emphasized that such cases are highly exceptional.

In conclusion, since the attempt theory harbours inconsistencies in terms of the theory of crime and would lead to an unjust application, it must be rejected. Instead, given that a causal link exists between the first act and the resulting outcome, an answer should be sought to the question of how this outcome can be imputed to the perpetrator from an objective and/or subjective perspective.

C. Adequacy Theory

b. Explanation of the Theory

The third theory developed under the concept of dolus generalis is the adequacy theory (Adäquanzurteil), which is based on the causality solution theory. According to this theory, the decisive factor in this type of case is that the deviation of the actual outcome from the anticipated causal link is merely insignificant.⁹⁰ This is because the real causal link is predictable according to general life experience. It is irrelevant whether the deviation is attributed to the perpetrator’s own act or to the influence of third parties or natural forces, since it is impossible for a human to have full knowledge of all conditions that bring about the result.⁹¹ It suffices that the perpetrator foresaw the causal link and that this foresight matches the actual causal link. In this assessment, the criterion that determines what is significant or insignificant is the foreseeability of the result according to general life experience. This theory is based on an adequacy assessment and a comparative evaluation of the actual causal link with the foreseen course of events. In the example of the attack on the victim followed by the submersion of the body, the deviation between reality and the perpetrator’s anticipation is neither extraordinary nor significant. Therefore, the perpetrator should be punished for a completed act of murder.⁹²

In the adequacy theory, since the act is considered a process, only the initial phase of this process is regarded as the perpetrator’s act in terms of criminal law. It is stated that the subsequent actions should be evaluated merely as a causal-mechanical link to the final outcome and that this link is significant only for the purpose of the adequacy assessment. It is emphasized that the subsequent actions, which are considered part of the perpetrator’s own act, bear no legal significance in this context.⁹³

According to the theory of insignificant deviation in the causal link, the German Federal Court of Justice (although it does not adopt the doctrinal concept of dolus generalis) concluded that there was an insignificant causal deviation from the intentionally committed act of killing and assessed the case as intentional homicide. This view, which dominates the German doctrine and is also upheld by the Federal Court of Justice,⁹⁴ holds that the occurrence of the death as a result of the second act should be considered an insignificant deviation in the causal process, and therefore, the perpetrator should be convicted of completed intentional homicide.⁹⁵ Thus, differing from the fundamental idea of the dolus generalis doctrine, it was not accepted that the perpetrator’s “general” intent to kill continued throughout the entire course of events; rather, it was acknowledged that the cause of death was not directly within the perpetrator’s contemplation, but this error was deemed insignificant.

“Because all details of an event cannot be predicted in advance, deviations from the anticipated course of events —provided that these deviations are foreseeable within the limits of general life experience— do not justify the conclusion that intent has been extinguished or that the act should be evaluated differently.”⁹⁶

In this case, it must be accepted that the offense of intentional homicide has been committed in such two-stage incidents.⁹⁷ Although the death occurs as a result of another act, the perpetrator has foreseen and accepted this possibility. The deviation in the causal process is insignificant and carries no legal significance.⁹⁸ It is irrelevant whether this deviation stems from the perpetrator’s own conduct or from the actions of third parties or natural forces.⁹⁹ However, for a completed offense to be punishable, the result must be objectively attributable to the perpetrator’s initial act. The debate here revolves around whether the risk arising from the perpetrator’s initial intentional act (e.g., death due to poisoning) materializes in the concrete result (e.g., the drowning of a poisoned, unconscious person).¹⁰⁰

According to the German Federal Court of Justice, although not directly, the perpetrator caused the fatal outcome indirectly at the moment they stuffed soil into the victim’s mouth with conditional intent (dolus eventualis). Accordingly, had it not been for the acts that indirectly paved the way for the victim’s death, the act completed with conditional intent (throwing into the cesspit) would not have occurred either.

“The perpetrator did not cause the death directly but rather indirectly. The perpetrator interpreted Ms. B.’s immobile state lying on the ground as an indication that she was dead and therefore threw the victim into the manure (cesspit) pit. This process, which directly impacted the victim’s death, would not have occurred without the perpetrator’s prior actions committed with conditional intent (dolus eventualis). Therefore, the perpetrator’s earlier actions are also regarded as causal for the death. In this context, the perpetrator caused the fatal outcome with conditional intent. Although the death occurred in a manner different from what the perpetrator had anticipated, the deviation between the actual causal chain and the causal chain envisioned by the perpetrator was extremely minor and insignificant from a legal standpoint.”¹⁰¹

Referring to its previous case law, the Court stated that there is no difference between direct intent (dolus directus) and conditional intent (dolus eventualis) in terms of causality, and ultimately ruled that the deviation in the causal chain, which developed differently from what was envisioned, was “legally insignificant.”¹⁰² This represents a partially consistent theoretical position while also offering a hybrid solution that appeals to the sense of justice. The Federal Court of Justice (BGH) rejected the previously dominant attempt theory (Versuchslehre), concluding that such a distinction between the first act and the result arising from the second act would arbitrarily deviate from the realities of life.¹⁰³

Accordingly, since the perpetrator caused (created a risk for) the later occurring death, which can be attributed to him, with the act he initially performed with the intent to kill, he should be punished for completed intentional homicide.¹⁰⁴ It makes no difference whether the perpetrator’s initial act was carried out with direct intent (dolus directus) or conditional intent (dolus eventualis).¹⁰⁵ The negligent offense realized by the second act remains in the background compared to the offense stemming from the intentional act.¹⁰⁶

The opinion expressed in the court’s decision changed the perspective on the previously dominant attempt theory. The Federal Court of Justice (BGH) had previously issued decisions in line with the attempt theory during the period when it was predominant.¹⁰⁷ Following this decision, the attempt solution was pushed into the background, and efforts were made to revise and develop the view presented in the decision,¹⁰⁸ aiming to fill the dogmatic gaps identified by the BGH.¹⁰⁹

b. Critique of the Theory

With its consistent decisions rendered in line with this view, the Federal Court of Justice (BGH) reaffirmed a solution to the problem that is consistent with both older legal traditions and relatively recent high court rulings. However, doubts regarding the solution presented in this decision have not been fully resolved, nor have the debates in the doctrine come to an end.¹¹⁰ The evaluation of such cases as a single completed intentional killing within the framework of the adequacy theory has been criticized from various perspectives.¹¹¹

This decision by the Federal Court of Justice (BGH) has been criticized by some scholars in German doctrine on the grounds that the result occurred not as a consequence of the risk created by the intentional act of killing, but rather as a result of another act and the will associated with it.¹¹² However, it should also be noted that some other scholars have expressed the opinion that the Federal Court’s decision is correct; according to them, if the result intended by the initial act occurs within the causal chain during a subsequent act, the existence of intent should still be accepted.¹¹³

Maiwald states that the Federal Court of Justice (BGH), in order to justify its view, essentially relied on the same arguments that have been known since v. Buri’s 1864 article and that have continuously provoked objections throughout the debates.¹¹⁴ The author emphasizes that proponents of the adequacy (suitability) theory overlook the fact that the act which actually led to death was carried out without the intent to kill.¹¹⁵ According to Maiwald, the particularity of dolus generalis cases lies in the fact that the forces effective after the perpetrator has “released” his action are not blind forces of nature but rather the perpetrator’s own conduct — that is, human will. In this context, if the adequacy theory is accepted, the following question must be answered: Are these actions controllable by the perpetrator and thus foreseeable? Proponents of the adequacy theory do not base their evaluation on this question. For them, the foreseeability of the causal chain alone is sufficient. However, dolus generalis cases can only attribute the resulting consequence to the perpetrator if it is assumed that the perpetrator could foresee the actual course (causal chain) of his action. A purely foreseeability-based assessment, ignoring controllability, is structurally inapplicable.¹¹⁶ The author has examined the issue from every angle and ultimately concludes that only the attempt theory fully explains the matter and that, therefore, a real concurrence (actual concurrence of offenses) must be applied between the two crimes arising in such a case.¹¹⁷

Among the authors criticizing this view, Hettinger asserts that proponents of the adequacy theory (suitability theory) lack a genuine argumentation and that these authors, instead of genuinely confronting the traditional structure of dolus generalis, try to reach the same conclusion through a specially crafted deviation theory — a situation he describes as a kind of “naming fraud.” According to the author, the doctrines of versari in re illicita, dolus indirectus, and dolus generalis are merely set aside in the adequacy theory, only to justify the desired outcome.¹¹⁸

One of the frequent criticisms of the adequacy theory is that it embodies a moral approach, which characterizes a “intent” (Gesinnung) based criminal law that takes into account not the intent in committing the crime, but merely the internal attitude of the perpetrator. Accordingly, Köhler argues that instead of a proper legal attribution, an assessment is made of a morally wrongful attitude such as wishing the death of others.¹¹⁹ Maiwald, in addition to this view, states that if only the intention of the perpetrator is considered, it becomes irrelevant whether the victim’s death occurred due to the initial assaultive act or merely as a result of subsequently throwing the body into the water.¹²⁰

Another scholar who supports the attempt theory, Sancinetti, argues that only the initial act can be attributed; even if a deviation in the causal link is accepted, this deviation is insignificant because the result occurred within foreseeable limits. He further states that this argument is generally structured on the ground that the event constitutes a single whole; however, according to him, this is not a genuine argument, since the fact that the event forms a whole does not prevent it from being evaluated as partly intentional and partly negligent.¹²¹

Roxin, however, takes a more moderate approach. According to Roxin, the Federal Court of Justice (BGH) reaches the conclusion that a completed homicide should be accepted by acknowledging that there is only an insignificant causal deviation between the perpetrator’s initial act and the resulting death. Roxin states that this view is widely accepted in the doctrine. Accordingly, the Court’s perspective, which evaluates dolus generalis —like similar previous cases— as merely a special form of causal deviation, should be appreciated. However, concluding that such deviation is always insignificant would be a “hasty approach”. For in such cases, it is often overlooked that the specific causal process might sometimes fall outside the scope of foreseeability; in such situations, it can be argued that the objective element of the crime does not exist. For example, if a person, thinking the victim is dead, loads them into a vehicle to bury them at a distant place, but the victim —who is actually only unconscious— dies in a traffic accident while being transported to the funeral site, it must be accepted that only an attempted crime has occurred due to the uncertainty of the causal process.¹²²

After stating that judicial decisions in dolus generalis cases accept the existence of an insignificant causal deviation, and that this is valid only when the perpetrator’s act carries the intent to kill, Roxin expressed that a minor limitation should be introduced. According to Roxin, the intent to kill must be uncorrected (i.e., not withdrawn or abandoned). If A initially intended to kill B but, upon seeing the victim motionless, feels remorse and abandons his intent, attempting to revive the victim (i.e., voluntary renunciation), but mistakenly assumes that his efforts have failed and then desperately buries the victim, thereby causing death, A would have abandoned his plan to kill during the attempt stage. In this case, the subsequent death cannot be regarded as the realization of the initially existing plan to kill and, therefore, cannot be associated with intent. However, in the majority of judicial decisions, the courts have ruled on cases where the intent to kill was not withdrawn, and thus the outcomes align with this view. Nevertheless, in its cesspit decision,¹²³ the Federal Court of Justice (BGH) recognizes that in dolus generalis situations, a completed homicide occurs even if the perpetrator acted with conditional intent (dolus eventualis), because, according to the BGH, the difference between the two types of intent and the deviation in the causal process from the perpetrator’s expectation is minor and legally insignificant. Roxin, however, argues that this difference must be opposed, asserting that it forms the very basis of the legal evaluation.¹²⁴ Building on this critique, Roxin has developed his own theory, which I will examine below.

A similar view has also been put forward by Walter. According to the author, a situation that must be distinguished from dolus generalis cases is where the perpetrator abandons their criminal intent before the second act, but through negligence or a mistake, causes the crime to be fully realized. Such cases can be referred to as “failed withdrawal” (verstolperter Rücktritt). This situation can arise in both completed and incomplete attempts. An example of a completed attempt would be a perpetrator injecting poison into the victim and then, wishing to save the victim with an antidote, mistakenly administering a higher dose of the active poison, thereby hastening the victim’s death. After completing the attempt, the perpetrator has abandoned the criminal intent, and during the withdrawal efforts, has caused the completion of the crime through negligence.¹²⁵ As noted above, Roxin also discusses the concept of “revised intention to complete” (revidierte Vollendungsabsicht) for such cases, arguing that there is no longer deliberate completion and thus concluding that there is a concurrence of an attempted crime and negligent completion.¹²⁶ Walter notes that the Federal Court of Justice (BGH) has indirectly adopted this solution,¹²⁷ but emphasizes that this situation concerns an incomplete attempt. In the case at hand, a dentist wanted to commit an act of revenge against his landlord by setting his office on fire. He poured gasoline in the office and struck a match but then changed his mind and tried to put the match back into the matchbox. However, the match accidentally ignited, and despite his immediate attempt to extinguish it, a small flame ignited the gasoline vapours, causing the entire office to catch fire. The Federal Court of Justice classified this only as negligent arson. According to the Court, at the moment the match was struck, the perpetrator no longer harboured criminal intent, and thus no longer had mens rea. It was established that the perpetrator had definitively abandoned his initial criminal intent at the stage of an incomplete attempt — namely, at a point where he believed he had not yet completed all the actions necessary to consummate the offense.¹²⁸ For the Federal Court, the brief time interval between the abandonment and the subsequent ignition was insignificant. Walter considers this approach consistent because, at the moment the perpetrator attempted to put the match back into the box, he was deemed to have voluntarily withdrawn under § 24(1) of the German Criminal Code (StGB), having abandoned the intent to continue committing the crime. The ignition of the match thereafter constituted a new and purely negligent offense.¹²⁹

D. The Perpetrator’s Realization of the Plan Theory

a. Explanation of the Theory

According to the Perpetrator’s Realization of the Plan Theory, Roxin asserts that even if the objective element of the offense of killing has been fulfilled in such cases, in order to determine the perpetrator’s intent, a distinction must be made as to whether the perpetrator intended the victim’s death or merely accepted it as a collateral consequence. If A intended to kill B, and after assuming that B was dead, buried the unconscious B, resulting in B’s death due to this second act, it must be evaluated —based on an objective assessment— that despite the deviation in the causal link, the situation constitutes the realization of the perpetrator’s plan. Within this plan, A intended to kill B and succeeded. Even though the course of events differed slightly, this difference is as insignificant in terms of the realization of the killing plan as other deviations in causality.¹³⁰ As can be seen, Roxin aimed to reach the conclusion previously arrived at by the Federal Court of Justice (BGH) through a more theoretically consistent approach, developing his own theory without distorting the criminal law doctrine, thereby pursuing a more “honest” path.

Roxin first points out that in unresolved issues of this kind, when choosing between two different solutions, one should not immediately conclude that “one is wrong.” This is because, starting from different premises, various internally consistent solutions can be developed. According to Roxin, in the set of cases defined under the term dolus generalis, the foreseeability of a deviation in the causal chain does not lead to a definitive conclusion; it only indicates that the consequence of the perpetrator’s second act is connected to the realization of the perpetrator’s plan and, thus, can be attributed to the perpetrator’s intent. However, if, at the time of committing the initial act, the perpetrator did not aim at the victim’s death but merely accepted the possibility of death —that is, acted with reckless intent (dolus eventualis)— the situation must be resolved differently.¹³¹ For instance, if a thief or rapist merely intended to render the victim unconscious and at most accepted the unwanted death as a risk, perhaps even attempted resuscitation but later wrongly assumed these efforts to be futile and buried the supposedly dead victim, resulting in death,¹³² this second act can no longer be considered the realization of the original plan, because it completely contradicts the perpetrator’s intent. In such a case, regarding the result of death, the perpetrator would not say, “I have achieved my goal anyway,” but would instead lament that everything went wrong due to a mistake. This evaluation is not only from the perpetrator’s perspective but normatively as well: it should be regarded not as the realization of a criminal plan but as an unfortunate accident, because the criminal plan only aimed at the success of the rape, and beyond that, sought to avoid death.¹³³ An example illustrating this assumption would be a situation where, after the initial act, the perpetrator realizes that the victim is still alive; in such a case, a perpetrator who acted with intent to kill would perform the subsequent lethal act without hesitation, whereas one who acted with reckless intent would not, as it would contradict their original aims. Thus, precisely for this reason, in dolus generalis cases, the existence of a completed crime should be assumed; otherwise, only an attempted crime should be acknowledged.¹³⁴ Consequently, according to Roxin, concretely, any act capable of bringing about the result initiates the causal process; what is decisive is whether there is still a foreseeable causal connection in the perpetrator’s plan.¹³⁵

This solution is largely consistent with the dominant theory concerning deviations in causality. For intentional crimes, two conditions are required to consider deviations insignificant: a) The deviation must remain within the limits of foreseeability. b) The deviation must not justify a different legal evaluation of the act. In dolus generalis cases, the principle of the realization of the perpetrator’s plan constitutes the decisive factor for attribution. Based on this principle, if the condition of foreseeability is satisfied, the result of the second act should be regarded as the realization of the perpetrator’s plan and should be assessed as a completed offense.¹³⁶

Roxin illustrates his view with the following example: If a football player aims at the lower left corner post, but the shot deflects and hits the upper corner post one meter higher and crosses the goal line, this is still considered by everyone to be an intentional goal (not an attempt), because the foreseeable causal deviation within the context of the realization of the plan does not change anything.¹³⁷

According to Roxin, the issue in such cases is whether the sequence of events can be objectively determined according to legal evaluation criteria in line with the perpetrator’s intent. This assessment of conformity is not cognitive but normative in nature, and it incorporates all causal and psychological findings arising from the course of events. However, these findings are subjected to a legal evaluation to examine whether the outcome is equivalent to what the perpetrator intended to achieve. At this point, he emphasizes that a teleological interpretation in criminal law is inevitable; otherwise, clear and definite solutions would be lacking. The analytical question, “How would the perpetrator have acted if they had realized their mistake in time?” yields quite clear results. Thus, it must be determined whether the plan has been objectively realized in practice, provided that the perpetrator’s intent to kill has not been altered by an attempt at voluntary withdrawal and remains unaffected by causal deviations. This ensures that the act actually committed by the perpetrator is made objectively visible.¹³⁸

With this view, Roxin supports the completion solution by arguing, on the one hand, that the second act —and thus the final result— constitutes a foreseeable course of events during the execution of the first act and, therefore, that the final result can be attributed to the first act as an insignificant causal deviation.¹³⁹ However, taking into account that foreseeability is generally a common feature of both intent and negligence,¹⁴⁰ and that the prevailing opinion supplements this concept with a test stipulating that “unforeseeable individual circumstances do not justify a different evaluation of the act,”¹⁴¹ he develops the concept of the “realization of the perpetrator’s plan” to resolve the question of whether the deviation from the intended course of events is imputable. This approach proposes that, in addition to the realization of the risk created intentionally, there must also be a finding that the “plan has been realized.”¹⁴² In cases of dolus generalis, the plan is deemed realized only if the perpetrator initially acted with intent to kill and did not revise this intent before the second act; otherwise (for example, if the perpetrator acted with conditional intent or even direct intent or purpose but later abandoned this purpose through contrary actions), the “perpetrator’s plan” is not realized. In such a case, the perpetrator’s first act remains at the stage of attempted intentional killing, and the realization of the result is attributed to the perpetrator’s likely negligent second act.¹⁴³

An approach closer to the solution provided by the theory of the realization of the perpetrator’s plan is the assumption that whether the result is to be considered as completion or attempt depends on whether the decision regarding the second act was made from the outset or only after the first act. According to this theory, if the intention to dispose of the victim arises only after the supposed killing, an attempted killing combined with negligent manslaughter under the principle of real concurrence of offenses would be present. This usually leads to the same conclusions advocated here; because a perpetrator who acts with intent to kill in the first act generally plans the second act from the beginning, as the disposal of the body will be necessary. In contrast, a perpetrator who acts only with conditional intent in terms of killing during the first act and hopes that the victim will survive usually will not have preplanned what to do with the body if the victim dies. However, both assumptions can be refuted in concrete cases, and relying on the timing of the decision to dispose of the body may be misleading; because the foreseeability of the second act generally does not depend on whether it was planned during the first act, nor is the evaluation of the second act as a realization of the danger created by the first act solely contingent on whether it was preplanned.¹⁴⁴

b. Critique of the Theory

Roxin’s view has been thoroughly criticized by the Argentinian criminal law scholar Sancinetti in an article published in the 2001 Festschrift for Roxin’s 70th birthday. According to Sancinetti, this distinction relies on a prerequisite of the traditional understanding: compared to direct intent (Absicht) pursued with a specific purpose, eventual intent (Eventualvorsatz) is considered legally inferior. Sancinetti argues that the doctrine of objective imputation (objektive Zurechnung) demonstrates that this is incorrect, since what matters for determining whether an act reduces an existing risk (and thus lacks criminal content) or replaces it with a lesser risk (and thus is lawful) is not the type of intent that creates the risk, but rather the objective dimension of the risk itself.¹⁴⁵ For example, a perpetrator aiming at a target with the intent to kill but using a defective weapon —thus having a low probability of hitting from a distance— creates less risk and commits a lesser wrong. By contrast, a perpetrator acting with eventual intent but shooting a person at close range with a large-calibre weapon creates a greater chance of success and commits a more serious wrong.¹⁴⁶ Attributing structural significance to the “will” element of intent requires understanding intent as “malice”, which, according to Sancinetti, must be rejected as a remnant of an outdated notion of intent-based criminal law; what matters is merely the perpetrator’s awareness of the relevant risk.¹⁴⁷ Moreover, a perpetrator who initially acts with a specific purpose may abandon this purpose in subsequent acts. For instance, if the perpetrator initially claimed confidence in achieving success with a single shot, they might lack motivation for a renewed attempt. Conversely, a perpetrator who initially acted with eventual intent might later find new motivation — for example, if the victim has been injured, the perpetrator might then seek to eliminate the victim as a witness by killing them.¹⁴⁸

According to Sancinetti, with a minor adjustment to Roxin’s view, these two objections can be overcome: what is decisive is not the type of intent with which the perpetrator initially acted, but whether the perpetrator would be motivated to commit a new (intentional) act upon recognizing the actual situation. The author notes that it is clear at various points in Roxin’s explanations that this distinction turns into a question about the perpetrator’s hypothetical behavior,¹⁴⁹ but that this means stepping outside the realm of act-based criminal law (Tatstrafrecht).¹⁵⁰ Moreover, it remains unclear why the type of intent should matter if the perpetrator acts without intent during the second phase, while it would not be important if the process continued without the perpetrator’s intervention or with some deviation. The explanation underlying the argument that the perpetrator “acted again” in dolus generalis situations¹⁵¹ obscures the fact that Roxin also regards the second phase as the “actual act of killing.” For when Roxin claims that the perpetrator initially acted intentionally and would have acted the same way even upon realizing the actual situation, he finds the attempt solution troubling. However, this cannot justify the claim that the final outcome is the realization of the initial risk.¹⁵² What must be realized is the danger encompassed by the perpetrator’s intent; if this risk materializes, an intentional completion occurs. Thus, the additional concept of a “realization of the plan” is misleading.¹⁵³

According to Sancinetti, the “correct solution” runs counter to intuitions, and particularly the more the conditions for “differentiated solutions” are met, the more intensely this contradiction to intuition arises: the perpetrator has foreseen the second act in their initial plan and created a risk; if this risk had not been altered by the perpetrator, success would have been achieved, and it would be accepted that the perpetrator acted with an unchanged intent to kill. Nevertheless, an intentionally completed offense does not exist. If there is no other argument to support this understanding, this intuitive solution must yield to dogmatic argumentation. This is the critical point for “reflective equilibrium” and also explains why dogmatic analysis and the dogmatic system must ultimately prevail over criminal policy.¹⁵⁴ In other words, according to the author, the solutions established by criminal dogmatics take precedence over criminal policy. Solutions deemed appropriate from a criminal policy perspective, despite contradicting criminal dogmatics, should not find application in criminal law. Liberation from a criminal law based on subjective views and its objective application can only be achieved in this way.

Roxin’s view has also been criticized for overlooking that this is a deviation issue. It makes no difference whether the so-called corpse is thrown into the water by a third party or by the perpetrator themselves; what matters is solely the adequacy assessment (Adäquanzurteil) and the realization of the risk created by the initial act. Therefore, if the perpetrator was determined from the outset to carry out the second act (the disposal of the supposed corpse), there is generally a completed intentional offense.¹⁵⁵ However, if the perpetrator decides to dispose of the body only after completing the act committed with the initial intent to kill, it can be said that the outcome is not inevitable and depends on the specific circumstances of the case; in such a situation, again only the adequacy assessment is decisive.¹⁵⁶ Significant deviations in the causal process can even be considered at the first stage.¹⁵⁷ For instance, if the perpetrator loads the person they believe to be dead into a vehicle to bury them in a remote place and the victim’s death occurs due to a traffic accident en route, this would be regarded merely as the realization of a general life risk, and due to the lack of objective attribution, there would be an attempted homicide and possibly negligent homicide.¹⁵⁸

E. Other Intermediate (Differentiated) Opinions

Among the differentiated opinions attempting to create an intermediate solution between the two extreme positions of completion and attempt, the first one is Welzel’s view. Generally advocating the completion solution, Welzel nonetheless accepts the real concurrence of offenses between attempted intentional homicide and negligent homicide in cases where the intent to dispose of the victim arises only after the supposed death. According to the author, what matters is whether the perpetrator foresaw from the outset the commission of the second act.¹⁵⁹

Stratenwerth¹⁶⁰ and Rudolphi¹⁶¹ proposed an intermediate view based on the perpetrator’s criminal plan. The authors justify their view by relying on the impact of the perpetrator’s foresight of the result in the initial plan regarding the second act. Accordingly, the focus should be on whether the later-occurring result of death constitutes the realization of the risk created by the perpetrator’s initial act. This reasoning would be valid if the perpetrator had foreseen the second act from the outset, but otherwise, it would remain doubtful. The authors leave unanswered the question of why it is necessary for the second act to have been foreseen at the beginning and for the result to stem from the risk created by the initial act. This approach assumes a situation in which the perpetrator deliberately uses themselves as a means for the later negligent act, resembling cases of actio libera in causa (alic). In any event, the second act, foreseen in the initial plan, aims to conceal an already completed crime and is not considered an alternative act of completion. Therefore, the foresight of the second phase in the initial plan remains ineffective in solving the problem.¹⁶²

Another intermediate solution proposed by Schroeder¹⁶³ suggests that if the initial act is concretely capable of causing the result—meaning the victim was already mortally wounded—then the act of burial or submersion accelerating the death would not negate the intent. However, if the initial act is not concretely capable of causing the result, then intent cannot be established.¹⁶⁴ According to this view, intent relies not on the causal link but on the idea that the act is concretely suitable to bring about the result.¹⁶⁵ Schroeder argues that what matters is not the degree of divergence between the actual process and the perpetrator’s envisioned process but the suitability of the act, as known to the perpetrator, for causing the result. This is because the subject matter of intent and negligence is not the main outlines of the causal process but the concrete suitability of the act for causing the result.¹⁶⁶ Applying this approach to the dolus generalis problem, Schroeder concludes that if the initial act was concretely suitable for causing the result —i.e., if the victim was already mortally wounded— the subsequent act of submerging the victim to hasten death would not exclude intent; however, conversely, if the initial act was not concretely suitable for causing the result, intent would be lacking.¹⁶⁷

Walter states that in dolus generalis cases, appropriate results regarding deviations in the causal flow can be obtained through the following test question: “If the perpetrator had known in advance how subsequent events would unfold when committing the initial act, would they have still carried out that act?” He notes that this criterion aligns with the fundamental principle that only the perpetrator’s mental conceptions are determinative for the assessment of intent, and that objective evaluations of what the perpetrator should or could have conceived are irrelevant. Furthermore, he asserts that this approach is dogmatically clear and reliably applicable, although he concedes that evidentiary issues may inevitably arise during the trial process. Walter also points out that the approach he proposes is close to, but not entirely consistent with, the view that differentiates between purpose (Absicht) and conditional intent (bedingter Vorsatz), because there may be unusual cases (atypical cases) where the perpetrator indeed intends to harm (kill) the victim but specifically avoids doing so in a particular manner or by a particular alternative means.¹⁶⁸

Another differentiated approach worth noting and discussing in this context is the natural unity of action solution advanced by the Spanish criminal law scholar Sanz Morán. Sanz Morán first clearly states that in such cases, there is a completed single intentional offense and, citing Karl Peters,¹⁶⁹ asserts that “the process from the beginning of the act to the result carries a unity of meaning (Sinneinheit)”. Thus, the issue lies in the distinction between a single offense and a plurality of offenses. Since there is no controversy that when the second act is committed with eventual intent, it constitutes a completed single intentional killing, it would be meaningless to speak of two different offenses when the second act results from possible negligence. The author states that the problem arises in the context of whether a single violation constitutes multiple offenses, emphasizing that a single violation can be concluded in different ways and all these ways can be reduced to a common idea. He asserts that the application of a single penal provision and the sanction prescribed therein are sufficient to reflect the total wrongful value of the act, and that any contrary approach would violate the principle of “non bis in idem”. He argues that in order to speak of a single offense, one could rely on the concepts of “apparent concurrence” and “natural unity of action”, with the latter meaning the evaluation of multiple homogeneous offenses as a single violation through a value-based synthesis. He contends that the existence of natural unity of action cannot be entirely disregarded in dolus generalis cases. However, beyond the applicability of the natural unity of action category to these cases, the author emphasizes that the distinction between a single offense and multiple offenses is a value-based matter rather than a logical-formal problem. This situation becomes evident in certain cases where it is necessary to decide whether there are multiple offenses —such as in natural unity of action and continued offenses— or in “preceding or subsequent connected acts”. According to the author, the material application of the non bis in idem principle (prohibition of double evaluation of a culpable element) plays a decisive role here. In this context, he expresses serious reservations against solutions to the dolus generalis problem that lead to the multiple evaluations of a single death incident.¹⁷⁰

V. The Treatment of the Subject in Turkish Doctrine

The theory of the “delayed realization of the result”, which is considered within the doctrine of dolus generalis by some authors in German scholarship, is referred to in Turkish doctrine as Weber intent (Weber kastı) and explained under this heading.¹⁷¹ Although it is not explicitly termed Weber intent in German doctrine, as detailed above, this issue is regarded as one of the problematic areas of intent (kast) and is evaluated within the dolus generalis theory as a particular type thereof.¹⁷²

It should be noted that the subject is not explained with such detail and significance in Turkish criminal law doctrine. Although the subject is presented under the heading of Weber intent (Weber kastı), thereby creating the impression that the classical dolus generalis solution is being adopted, as explained in detail above, the issue is actually resolved within the framework of the adequacy (Adäquanzurteil) approach of the German Federal Court. In Turkish doctrine, it is stated that in dolus generalis cases, it is not necessary to determine all the details of the causal link and the exact moment of the result, such as the precise moment of death; rather, it is deemed necessary and sufficient that the course of events generally conforms to the perpetrator’s long- or short-term plan or intention. Accordingly, a minor deviation between the act and the result does not affect the intent (mens rea). Based on these explanations, it is stated that an evaluation according to general life experience can be utilized to distinguish whether the deviation in the causal link is significant, and that deviations which are foreseeable according to general life experience and which do not necessitate a different assessment of the perpetrator’s act are considered insignificant. For example, if a perpetrator strikes the victim’s head with a hammer, assumes the victim is dead, and throws the body off a bridge to dispose of it, but the victim actually dies by striking the pier of the bridge or drowning, it is accepted that Weber intent is present, and it is asserted that the deviation is insignificant and that the perpetrator should be punished for completed intentional homicide.¹⁷³ Accordingly, in cases where the intent to kill and the deviation do not occur simultaneously —for instance, where the perpetrator, after shooting the intended victim, assumes that the victim is dead and buries the victim alive, leading to death by suffocation— it is accepted that the perpetrator aimed to realize a specific result or at least considered it possible. In such a case, following the shooting, the delayed realization of the result occurs with the second act, and it is accepted that the legal definition of the offense is fulfilled.¹⁷⁴

In the doctrine of Turkish criminal law, Önder, Artuk/Gökçen et al., and Demirbaş¹⁷⁵ examine this issue under the heading of Weber intent (Weber kastı) and assert that the deviation is insignificant, thereby holding the perpetrator liable for completed intentional homicide. Hakeri, without giving a specific designation, provides a solution based on the theories of attempt and completion and states that he supports the completion theory.¹⁷⁶ Apayadın, in his monographic work on intent, refers to Önder, reiterates the fundamental twofold distinction in this field using the same method, and expresses that the “error” in the causal link is not significant.¹⁷⁷ Dönmezer/Erman assert that the initial intent is sufficient for the crime to be considered intentional and that even if the perpetrator regrets at the moment the result occurs, their criminal liability based on intent persists.¹⁷⁸ Öztürk/Erdem,¹⁷⁹ Özbek/Doğan/Bacaksız,¹⁸⁰ Koca/Üzülmez,¹⁸¹ and Özen¹⁸² address the issue in the context of deviation in the causal link and argue that this deviation is insignificant, thereby concluding that the perpetrator should be punished for completed intentional homicide. At this point, İçel emphasizes that an error or deviation in causality does not transform liability based on intent into liability based on negligence, as intent does not need to encompass the causal link.¹⁸³ In summary, the prevailing opinion in the doctrine of Türkiye is also parallel to the solution and reasoning of the German Federal Court (BGH).

As can be seen, although the term Weber intent (Weber kastı) is used in Turkish criminal law, creating the impression that the dolus generalis doctrine has been adopted, in reality, the solution based on adequacy (Adäquanzurteil) developed by the German Federal Court (BGH) is accepted. However, as I have repeatedly noted throughout this article, the view initially proposed by Weber has lost its validity through the decisions of the BGH. Particularly among the authors who examine the subject under the heading of Weber intent, there is a misleading impression —as if Weber had introduced the currently prevailing view— even though they are, in fact, referring to the current doctrine. Furthermore, the historical development and transformation process between Weber’s view and the BGH’s modern approach is not mentioned, not even in summary.

This view, which is only briefly mentioned in the criminal law doctrine in Türkiye, has never been addressed in judicial decisions. It is, however, not surprising that a subject which has not been thoroughly discussed in the doctrine has not found its way into judicial rulings.

VI. My Opinion

First and foremost, it should be stated that I believe all of the views explained above have both internal inconsistencies and highly coherent aspects. While each perspective has points open to criticism, it also presents reasonably acceptable justifications. In my view, there is, in fact, no single solution that can be universally applied to all cases of dolus generalis. However, the predominant solution method varies depending on the perspective from which the case is approached. For instance, if it is emphasized that the second act would not have been committed if the perpetrator had not performed the first act, and that a fair outcome requires the process to be considered as a whole, the completion theory is adopted; if the concurrence between the objective and subjective elements of the offense, that is, simultaneity, is accepted, the attempt theory is preferred; and if the death result occurring through the second act is considered an insignificant deviation in the causal process, the adequacy theory is embraced.

The conclusion and opinion I have drawn from my study are as follows: Although the Weber intent or the doctrine of dolus generalis, which is well-established in Turkish legal scholarship, seems to yield a fair outcome, I believe it contains theoretical inconsistencies that should not be overlooked. While the perpetrator’s general desire may indeed be directed toward killing the victim, the two acts in question, their consequences, the deviation in the causal link, and the mental elements of these acts are significantly different from each other. This solution equates the pre-intent desire with intent itself and, albeit implicitly, carries the danger of drifting towards an intent-based criminal law (Gesinnungsstrafrecht). Furthermore, treating intent in this manner could ultimately lead to the recognition of only one form of intent — that of eventual intent (dolus eventualis).

On the other hand, although the acceptance of the attempt theory would be theoretically more accurate and consistent, it would result in an outcome that does not appeal to the sense of justice. In this case, since the perpetrator has committed two different acts resulting in two different crimes, they would need to be punished separately for each crime (attempted intentional homicide and negligent homicide). This would mean that the perpetrator, despite achieving the outcome they intended, would be rewarded for the misfortune in the causal chain. One of the important functions of criminal law is to ensure a distinction between misfortune and wrongdoing. In my view, this solution fails to make such a distinction in a sound manner.

Although the solution offered by the Federal Court of Justice (BGH) appears to be more reasonable and aims to find a middle ground, it is not possible to claim that the deviation in causality is insignificant in every case. The deviation in the causal link is not always minor. In cases of significant deviation, it cannot be said that the causal process is suitable for leading to the result. Therefore, this solution cannot be applied universally.

Although the differentiated views enrich the dogmatic discussion from a theoretical perspective, they do not offer practical solutions that can be applied in every situation.

In terms of the theory advanced by Roxin regarding the perpetrator’s realization of his plan, it appears reasonable to hold the perpetrator liable for the offense of intentional killing committed through two separate acts, on the grounds that the actions carried out by the perpetrator are parts of the plan and that the perpetrator, by executing his acts within this plan, accomplishes the intended result. However, it should also be noted that this theory carries certain risks due to the requirement that all the results must be part of the plan. According to Roxin, the perpetrator should not be held responsible for actions and outcomes that are not included in the plan. Yet, it is not possible for the perpetrator to possess eventual intent (dolus eventualis) with respect to a matter that is not part of the plan; in such a case, there would be no dolus generalis situation. Dolus generalis covers situations where the intended result has not been realized through the initial act; the absence of the result from the perpetrator’s original plan indicates that there is no eventual intent. When we examine the examples of theft and rape given by Roxin to justify his theory, as well as the subsequent explanations, it is understood that the perpetrator is practically considered to have realized an unforeseen outcome through the second act and therefore cannot be held liable for the realized result. However, in such a situation, there is no eventual intent to speak of, and thus, no dolus generalis issue arises. Moreover, in Roxin’s examples, the perpetrator’s primary objective (theft) is different from the incidental outcome (death) that occurred, and the outcome (death) is presented as something that was not foreseen, anticipated, or included in the plan by the perpetrator. Nevertheless, it is simultaneously stated that the perpetrator had eventual intent regarding the incidental outcome. It is quite difficult to argue that there is eventual intent concerning outcomes that are not foreseen, not anticipated, or not included in the perpetrator’s plan. Therefore, while I fully agree with Roxin’s view that exceptional cases may exist, I believe that the examples provided, and the accompanying explanations contain inconsistencies and do not fully reflect dolus generalis cases.

After the criticisms I have provided for each view above, it should be stated that the application of the decisions and explanations of the Federal Court of Justice (BGH) appears to be the most reasonable solution method for the vast majority of problems created by dolus generalis cases. Nevertheless, it must not be overlooked that the BGH, by aiming for a practical solution, tends to deem every deviation insignificant, thus somewhat glossing over the theoretical debates on this issue. Indeed, in the vast majority of dolus generalis cases, the deviation may be insignificant, but in some cases that are not accounted for, the deviation may be truly noteworthy. In such situations, the deviation may be so substantial that the natural connection between the two acts committed by the perpetrator is severed, and thus I believe that the theory of attempt could be applied. For example, if the perpetrator strikes the victim with intent to kill and, believing the victim to be dead, places the body on the back seat of his car in order to dispose of it, but while driving to the seaside for disposal, the victim dies in a car accident en route, there is a significant deviation. Although a completed offense technically exists in this case, the disconnection between the perpetrator’s actions would weaken the argument that the perpetrator deserves a very severe penalty, and thus the injustice resulting from the application of the theory of attempt could be partially alleviated.

In conclusion, I would like to express that the views proposed to resolve the problems of dolus generalis do not, on their own, offer a complete and definitive solution; that examples can be produced to both affirm and refute each view; and thus, this issue will continue to remain one of the controversial topics within the theory of crime.

VII. Reverse Dolus Generalis Cases

Another point I would like to address in this study is the occurrence of the exact opposite of the aforementioned examples, namely the situation of reverse dolus generalis. In such a case, the perpetrator plans to achieve a specific result, such as causing death, through certain actions; however, the result is achieved by one of the initial actions or even directly by the first act — that is, there is a situation of “early completion.”¹⁸⁴ For example, if a person, after slapping their spouse to the point of unconsciousness, intends to kill them by throwing them from a high-floor apartment, but the spouse dies as a result of the slap before being thrown from the height, what should be the conclusion?¹⁸⁵ In this case, the perpetrator did not foresee the result when performing the initial act, nor did they manifest an intention toward it, yet they achieved the result they had planned to accomplish with the second act through the first act.

In a decision rendered by the German Imperial Court,¹⁸⁶ the perpetrator who wanted to kill his wife attempted to give the impression that she had jumped off the train by herself, thereby disguising his actions as a suicide. To achieve this, he struck his wife’s head hard with a wrench and subsequently pushed her onto the tracks. The court could not definitively determine whether the death resulted from the blow inflicted by the wrench causing a concussion or from the impact sustained when she fell onto the tracks. In this situation, the Court stated that it was irrelevant whether the perpetrator had struck the victim’s head merely to impair her cognitive functions or with the intent to kill her at that moment; even if the blow had been delivered solely with the intent to injure and the fatal result occurred at that point, this would not be significant in terms of establishing intent.

In resolving this case, the court employed the following example: If a person intends to kill another with a hammer, but does not strike the first blow with the intent to kill, and his actual aim is to strike the victim multiple times until death ensues —planning that death will occur only after several blows— yet death occurs after the first strike, it is still considered that the killing act was committed with intent. What differentiates this situation from the classical dolus generalis intent is that unlike in the classical dolus generalis scenario, the perpetrator’s initial act is not aimed at killing. Rather, the intent to kill is directed at the second or subsequent acts, with the perpetrator expecting that the first blow will not be sufficient and that further blows will be necessary.¹⁸⁷ Indeed, in the hammer example given by the court, if the perpetrator intended to kill the victim after perhaps twenty blows but death occurred after the first strike, and he then ceased his actions, he would not be held responsible for the aggravated form of killing characterized by cruelty. It is clear that the perpetrator’s pre-legal motive or drive in his internal world has no bearing on the commission of the act or the realization of the result; such an intention remains outside the realm of punishable conduct.

More recently, German courts have dealt with two other situations that have been the subject of special scrutiny by contemporary doctrine:¹⁸⁸ In the first case, the perpetrator aimed to incapacitate the victim and then planned to take the victim approximately 100 kilometers away to kill him; however, the victim died during the initial acts (binding, gagging, and administering drugs) aimed at rendering him helpless. In the second case, the perpetrators intended to knock the victim unconscious by strangling his neck and then planned to cause death by injecting air into his veins, but the victim had already died due to the initial assault.

These incidents share a common characteristic with dolus generalis cases: the perpetrator aims to achieve a specific typical result (death in the cases mentioned), and this result occurs through a causal process that does not conform to the perpetrator’s planned sequence. However, in dolus generalis cases, the initial act intended to cause death does not result in death; rather, death occurs through a subsequent act. In contrast, in cases of early completion, the first act —which was not intended to cause the result— nonetheless brings about the result.

In this case, it must be assumed that the death resulting from the initial act is objectively attributable, meaning it was foreseeable for the perpetrator. Based on this assumption, the question arises whether the earlier occurrence of the result — namely, the fact that the perpetrator’s initial act caused death earlier than planned —should be considered a legally insignificant causal deviation and whether this can be regarded as a completed homicide.¹⁸⁹

The prevailing opinion in the doctrine is that this constitutes a deviation from the causal course foreseen by the perpetrator. Such deviations are considered insignificant, and it is accepted that a completed intentional offense exists in this situation.¹⁹⁰

Puppe, aligning with the court, argues that any act rendering the victim unconscious, based on the view that a blow to the head with a wrench is a typical method of killing, should be considered within the concept of “intentional danger”¹⁹¹ and that the perpetrator should be punished for a completed offense. According to her, the perpetrator had already initiated the realization of the killing result by applying a method capable of achieving it by rendering the victim defenceless. Furthermore, since it cannot be expected for the perpetrator to foresee every stage of the causal development, there is no factor that would affect causality.¹⁹²

Schroeder argues that in such cases the perpetrator lacks intent to complete the offense. According to Schroeder, if the result occurs during an act that constitutes an incomplete attempt, it must be accepted that the perpetrator did not have knowledge that their behaviour would concretely cause the result. In this case, there is a mistake regarding the principal act and an incomplete attempt, where the perpetrator could easily renounce the crime voluntarily and, as a result, would likely be liable for a negligent offense.¹⁹³

Sternberg-Lieben and Schuster, however, argue that this view is unacceptable because even an incomplete attempt stems from the perpetrator’s intent to realize the elements of the crime, and the German legislator treats incomplete and complete attempts equally in terms of wrongfulness and culpability.¹⁹⁴ For example, if a perpetrator believes that it would take ten doses of a particular poison to cause a person’s death, according to Schroeder, the perpetrator has not yet formed the intent to complete the crime — that is, the idea that they have done everything necessary for the result to occur. However, Sternberg-Lieben and Schuster assert that even in cases of early completion, it should still be considered an intentional act. Therefore, where there is an insignificant deviation in the foreseen causal link, the perpetrator should be punished for a completed intentional offense. Voluntary renunciation can only be considered in exceptional cases where the result cannot be attributed to the perpetrator’s conduct.¹⁹⁵

However, if the result occurs contrary to the perpetrator’s foresight during the preparatory stage, this is considered a significant deviation. In such a case, the perpetrator may be held liable for a negligent offense. This is because if a preparatory act triggers the occurrence of the result, a completed intentional offense cannot be recognized.¹⁹⁶ For instance, this would apply in a situation where a rifle accidentally discharges while being cleaned, or in a kidnapping case where the victim, intended to be killed after several intervening steps elsewhere, dies prematurely. The Federal Court of Justice (BGH) similarly states that during the preparatory phase, a “legally valid” intent does not yet exist and that the initial act should only be evaluated as negligent homicide (or bodily harm resulting in death in the kidnapping case). The reason that a definitive decision to commit a specific offense during the preparatory acts does not yet constitute “intent” is that the result must stem from an intentionally committed act that constitutes the elements of the offense, and such an act has not yet occurred during the preparatory phase.¹⁹⁷

In resolving this issue, it could be argued that the danger created according to the perpetrator’s decision should be considered within the scope of intent even during the preparatory stage or attempt, as it is irrelevant which version of the act results in the outcome because all versions can be attributed to the same act of creating a risk of killing. However, this should not be accepted according to the act-based criminal law approach (Tatstrafrecht), as it would excessively broaden the scope of punishment into the area of preparatory acts.

Although there is a tendency to resolve both dolus generalis and reversed dolus generalis situations in the same way, the legal justification can only be the same for both cases when considered from the perspective of the completion solution. In this way, in both situations, the deviation in the causal link is regarded as insignificant. However, in the attempt solution, the reason for rejecting completion is that in the case of dolus generalis, the original risk does not materialize in the result caused by the second act, while in the case of early completion, the perpetrator has not yet developed a “perfect” intent (completion of the attempt).¹⁹⁸

On the other hand, some authors who advocate the attempt solution for dolus generalis do not find the same solution suitable for “early completion”.¹⁹⁹ In fact, compared to dolus generalis, reaching the attempt solution is easier in cases of early completion, because in dolus generalis, the perpetrator has already foreseen that what they have done could cause the result; therefore, in situations where the initial risk has not been truly replaced by a second risk and only the accompanying circumstances have changed, a completed intentional act can be recognized. In contrast, in reversed dolus generalis, the perpetrator has not yet considered the possibility of completion.²⁰⁰ The fact that authors who find the attempt solution appropriate for dolus generalis do not find it appropriate for reversed dolus generalis seems to us an interesting approach.

I am of the opinion that approaching the matter from the perspective of the attempt theory in cases of early completion would most likely lead to an erroneous conclusion. In such a case, the perpetrator has already commenced their act with suitable means and methods to achieve the result they intended (and planned to achieve after a few subsequent actions). Since the perpetrator has begun an act aimed at the result, it is clear that they possess the requisite intent. The perpetrator merely did not anticipate that the result could occur through the initial actions or means. Thus, there is no issue with the perpetrator’s act or intent; there is only a minor deviation in the causal process. To regard this deviation as significant and to argue that the perpetrator could have voluntarily renounced the crime —and therefore should only be punished for an attempted crime— would amount to rewarding a perpetrator who has clearly demonstrated a criminal intent. This, however, would conflict with the concept of justice (not with mere intuition) as interpreted teleologically in law. Conversely, considering the deviation insignificant is dogmatically consistent because it is grounded in the fundamental principles of criminal theory.

Therefore, in the case of reverse dolus generalis, a dogmatic problem or debate does not arise as intensively as it does in classical dolus generalis cases. Indeed, whether one examines early completion cases based on the adequacy theor or the theory of the realization of the perpetrator’s plan, the same conclusion is reached. This outcome also demonstrates that the attempt solution does not yield a theoretically correct result in this context. When differentiated solutions are applied to early completion cases, the conclusion I have reached is likewise confirmed.

VIII. Conclusion

The significance of dolus generalis cases lies in the fact that, for two centuries, they have been at the center of discussions on the most important dogmatic issues of criminal law theory, and these debates will continue into the future. Thus, many components of criminal law theory are positioned, harmonized, and justified in an eclectic manner over a single subject, producing logical and coherent practical outcomes. In this way, the components of criminal law theory are, in effect, subjected to a consistency test. Cases such as the manure pit and similar incidents are extremely important in providing a foundation for such discussions.

The discussions conducted on this topic and type of case demonstrate to us this richness and diversity. In this context, the primary issue debated is whether the final outcome (death) can be objectively attributed to the perpetrator’s initial act— namely, the elements of objective typicity are examined. This discussion may take place within the framework of attribution theories, such as the theory of adequate causation (assessing whether the perpetrator’s act was suitable to cause the death), or within the context of the theory of conditions explaining the causal link and its limitation through the theory of objective attribution (whether the risk created by the initial act is accepted as materializing into the death). Regarding the subjective elements of typicity, the issue is the necessary alignment between the intent (mens rea) and the objective elements of typicity (actus reus). Therefore, compatibility problems arise, evaluated in terms of whether the causal link between the acts and the result is considered “significant” or “insignificant.”²⁰¹

Most of the proposals aimed at resolving this issue focus on the fundamental aspects of criminal law theory mentioned above. However, the proposals are not limited to these, and the issue has also been approached by reference to other elements of criminal law theory. For instance, it has often been argued that the completed crime solution provides a fairer system of punishment based on considerations of justice, or the solution has been subjected to moral critique for its implicit proximity to intention-based criminal law. Moreover, from time to time — though less frequently today— the concept of dolus generalis has been explained by drawing analogies to other dogmatic categories such as indirect perpetration or actiones liberae in causa. Some proposals referred to as differentiated solutions attempt to create a solution by adding differentiating elements to the adequacy theory, such as a special assessment of the subjective element of the initial act (e.g., the distinction between direct intent and conditional intent, or whether there was an intention to initiate the second act). Finally, certain approaches incorporate complex issues like the unity and plurality of acts or crimes into the discussion.²⁰²

At this point, a distinction must be made between two analyses: a) Whether the final result can be objectively attributed to the initial act. b) If it can be attributed, whether this attribution is also valid in terms of intent. In this context, it is clear that the initial act is causally relevant for the final result. According to the “conditio sine qua non” formula of the condition theory, if the perpetrator had not performed this initial act deliberately, there would have been no need for the second act; therefore, eliminating the initial act would also eliminate the final result.²⁰³ Thus, from the perspective of objective elements of the offense, there is no problem in terms of the causal link between the initial act and the final result that occurred.

However, the more contentious issue —and the examination that must be conducted in terms of determining the perpetrator’s responsibility— is whether, despite the initial act causing the result, this result can also be objectively attributed to that act. There is no doubt that the initial act created a legally unacceptable risk. The debate centers on whether the death can be considered as the realization of the risk created by the initial act.²⁰⁴ Sowada states that, within the framework of general foreseeability, it would be consistent with the ordinary experiences of life for a person without specialized medical knowledge to confuse the victim’s condition with unconsciousness and to kill the victim through the act of (hypothetically) disposing of the body. Thus, the occurrence of the death should be regarded, in the context of the initial act, as the wrongful result, and not as a mere accident.²⁰⁵

In parallel with this view, those who refer to the perpetrator’s plan in carrying out the initial act or to the intention to carry out the second act during the initial act advance similar arguments. In this context, Roxin responds to the thesis defended by Sancinetti — that the death is the realization of a different risk rather than the initial one. Roxin argues that the view of “the creation of two different risks” is not convincing. He maintains that, for the attribution of a deliberately committed and completed crime to the perpetrator, it is in any case necessary that the death occurring through the second act must have been foreseeable in advance. On the other hand, if the prohibited risk of the second act was already inherent in the first act, Roxin states that it is not entirely clear why the result should be regarded as the realization of a completely different risk rather than the initial one.²⁰⁶

From this point of view, regarding the subjective element of the offense, the following question must be answered: “Can the death of the victim be subjectively attributed to the perpetrator as an intentional consequence of his action?”²⁰⁷ German case law and a portion of the doctrine do not see any issue in such situations. According to them, there is merely an insignificant deviation from the causal sequence envisioned by the perpetrator. Accordingly, the perpetrator’s intent must cover the causal sequence, but it is sufficient that he broadly envisions the connection between his action and the result. Since it is impossible to know the real causal sequence exactly, it cannot be expected that the perpetrator should know it either.²⁰⁸ The counterargument to this view asserts that it contradicts the simultaneity principle, which requires that intent must exist at the time of the act. In this situation, the act that causes the victim’s death would have been carried out without intent to kill. In response to this objection, Roxin argues that it is not necessary for the intent to exist throughout the entire process but only at the moment when the perpetrator relinquishes control over the causal link. According to Roxin, in dolus generalis cases, the death of the victim is attributed to the perpetrator as an appropriate consequence of the first act encompassed by his intent. This inference is sufficient for the result to be attributed to the perpetrator’s intent because the result is accepted as the realization of the perpetrator’s plan.²⁰⁹

Within the framework where the discussion is thus limited, the divergence of opinions essentially stems from the fact that proponents of the completed crime theory focus on the “volitional element” of intent present in the initial act. This view considers it unnecessary to require precise knowledge of the risk that leads to the victim’s death. In contrast, proponents of the attempt solution emphasize the cognitive aspect of intent more strictly and argue that the perpetrator’s conception must also include precise knowledge of the risk causing the victim’s death. Ultimately, which view is adopted depends on whether priority is given to the cognitive or volitional aspect of intent.²¹⁰ I believe that in criminal theory, emphasis should be placed on the cognitive aspect of the perpetrator —that is, the knowing element of intent— and I consider this element to carry more weight than the volitional element, or in other words, the wanting element within intent. However, I also believe that knowledge alone is insufficient, especially in resolving borderline cases like the ones discussed in this article (similar to the distinction between dolus eventualis and conscious negligence). I think it is unreasonable to expect a person to be fully aware of all the processes they are involved in or to possess near-certain knowledge of these processes.²¹¹

In my opinion, there is no single solution that can be applied to all cases of dolus generalis. However, the predominant method of resolution varies depending on the perspective from which the situation is approached. For instance, if it is emphasized that the process should be considered as a whole because the perpetrator would not have committed the second act without performing the first, in order to achieve a just outcome, the completion theory can be adopted; if the simultaneity of the objective and subjective elements of the offense is accepted, then the attempt theory can be applied; and if the occurrence of the fatal outcome through the second act is regarded as an insignificant deviation in the causal chain, then the adequacy theory (Adäquanztheorie) can be preferred.

Proponents of the completed offense theory criticize the supporters of the attempt solution by arguing that “the handling of causal deviations is not a matter of logic, but rather an issue of policy or another form of value judgment”. In contrast, supporters of the attempt theory claim that their opponents are engaging in a form of “decisionism”. However, it must be acknowledged that resolving such situations inevitably involves a certain degree of decisionism.²¹²

This discussion brings us to the fundamental question of what the essential of the law truly belongs to. Without answering this question, it is impossible to resolve the concrete issue at hand. In other words, is law purely a field of science, or is it an area concerned with values? To answer this, we must first address the question: What is “right,” the foundation of law? Can the concept of a right be concretely defined and pointed to, like an apple, a book, a strand of hair, or a computer? It is not possible to define and point to the concept of a right in such concrete terms. Thus, the concept of a right is essentially a value judgment. A person’s possession of a right, or the violation of their right, depends on the legal system having recognized that right for the individual. This act of recognition necessarily involves a value judgment about something. Someone, or a group (such as the legislator), makes a value judgment in granting a particular individual or group the recognition of something as a right. For example, when I say, “The intellectual property rights of this article belong to the author,” I am not referring to the physical existence of the article as printed on paper or displayed digitally, but rather to a recognized value judgment that is protected by law — the legislator deems intellectual creation as a value that must and can be protected as a property right. Thus, the foundation of law is intrinsically tied to the notion of rights and, consequently, to value judgments.

From this perspective, it can be said that law is formally —that is, in terms of its structure— a natural science when it comes to external elements such as the hierarchy of norms, the enactment of laws, the principles of no crime and no punishment without law, the prohibition of retroactive legislation, or the imposition of life imprisonment for murder. However, regarding its substantive content — prohibitions like “you shall not kill,” “you shall not steal,” “you shall not commit sexual assault,” and imperatives such as “you shall protect and care for your parents and children” or “you shall perform your duties on time and properly”— law always involves value judgments, and in this respect, it can be considered a cultural science. These prohibitions and imperatives are formulated based on certain value judgments. Thus, in terms of its content, law, being based on values, may sometimes act as a “decision-maker,” embodying and expressing value judgments. Dolus generalis cases are among those situations where such value judgments manifest themselves and where a decision must be made based on an evaluation.

In my view, applying the decisions and reasoning of the BGH (Federal Court of Justice) appears to be the most reasonable method for resolving the vast majority of the problems created by dolus generalis cases. However, it should not be overlooked that, because the BGH aims for a practical solution, it tends to gloss over the theoretical debates by deeming every deviation insignificant. Indeed, in most dolus generalis cases, the deviation may be minor, but in a small number of cases, the deviation could actually be significant and worthy of consideration. In such instances, the deviation may be so substantial that the natural connection between the two actions carried out by the perpetrator could break, and at that point, the attempt theory might become relevant.

In conclusion, I believe that the views put forward to resolve the issues surrounding dolus generalis do not, on their own, offer a complete and definitive solution. For each view, examples can be produced both supporting and refuting it.

Thus, this matter will likely continue to remain one of the controversial topics within criminal law theory.

REFERENCES

Apaydın, Cengiz, Ceza Hukukunda Doğrudan Kast, Olası Kast, Basit Taksir ve Bilinçli Taksir Kavramları, no place of publication, date, or publisher.

Artuk, Mehmet Emin /Ahmet Gökcen/Mehmet Emin Alşahin/Kerim Çakır, Ceza Hukuku Genel Hükümler, 13. Baskı, Adalet Yayınevi, Ankara, 2019.

Backmann, Leonhard E., “Grundfälle zum strafrechtlichen Irrtum”, Juristische Schulung (JuS), 1972.

Binding, Karl, Handbuch des Strafrechts, Band I, Duncker & Humbold, Leipzig, 1885.

Demirbaş, Timur, Ceza Hukuku Genel Hükümler, 15. Baskı, Seçkin Yayıncılık, Ankara, 2020.

Dönmezer, Sulhi /Sahir Erman, Nazari ve Tatbiki Ceza Hukuku Genel Kısım, C.

II, 12. Baskı, Beta Yayıncılık, İstanbul, 1999.

Dülger, Murat Volkan, Ceza Hukuku Genel Hükümler, 2. Baskı, Seçkin Yayıncılık, Ankara, 2023.

Engisch, Karl, Untersuchungen über Vorsatz und Fahrlässigkeit im Strafrecht, Scientia Verlag, Aalen, 1964.

Eser, Albin/Björn Burkhardt, Strafrecht Bd.I, Schwerpunkt Allgemeine Verbrechenselemente, 4. Auflage, C.H.Beck, München, 1992.

Finger, August, Lehrbuch des Deutschen Strafrechts, Carl Heymans Verlag, Berlin, 1904.

Fischer, Thomas, Strafgesetzbuch: StGB, 72. Auflage, C.H. Beck, München, 2025.

Frank, Reinhard, Das Strafgesetzbuch für das Deutsche Reich nebst dem Einführungsgesetz, J.C.B. Mohr, Tübingen, 1931.

Freund, Georg, Strafrecht Allgemeiner Teil: Personale Straftatlehre, 2. Auflage, Springer, Berlin Heidelberg, 2009.

Frisch, Wolfgang, Tatbestandsmässiges Verhalten und Zurechnung des Erfolgs, C.F. Müller Verlag, Heidelberg, 1988.

Geyer, August, “Zur Lehre vom dolus generalis und vom Kausalzusammenhang”, Goltdammer's Archiv für Strafrecht (GA), Band 13, 1865.

Graf zu Dohna, Alexander, Der Aufbau der Verbrechenslehre, 4. Auflage, Ludwig Röhrscheid, Bonn, 1950.

Gropp, Walter, Strafrecht Allgemenier Teil, 4. Auflage, Heidelberg, Springer, 2015.

Hakeri, Hakan, Ceza Hukuku Genel Hükümler, 24. Baskı, Adalet Yayınevi, Ankara, 2021.

Heinrich, Bernd, Ceza Hukuku Genel Kısım, C. II, Adalet Yayınevi, Ankara, 2015.

Hettinger, Michael, “Der sog. dolus generalis: Sonderfall eines „Irrtums über den Kausalverlauf”?”, Festschrift für Günter Spendel zum 70. Geburtstag, Walter de Gruyter, (Hrsg) Manfred Seebode, Berlin New York, 1992.

Herzberg, Rolf Dietrich, “Aberratio ictus und abweichender Tatverlauf”, Zeitschrift für die gesamte Strafrechtswissenschaft (ZStW), Vol. 85, Issue 4, 1973.

Hruschka, Joachim, Strafrecht nach logisch-analytischer Methode: Systematisch entwickelte Fälle mit Lösungen zum Allgemeinen Teil, 2. Auflage, Walter de Gruyter, Berlin New York, 1988.

Hruschka, Joachim, “Die Herbeiführung eines Erfolges durch einen von zwei Akten bei eindeutigen und bei mehrdeutigen Tatsachenfeststellung”, Juristische Schulung (JuS), 1982.

İçel, Kayıhan, Ceza Hukuku Genel Hükümler, 6. Baskı, Beta Yayıncılık, İstanbul, 2021.

İçel, Kayıhan/Füsun Sokullu-Akıncı/İzzet Özgenç/Adem Sözüer/Fatih S.

Mahmutoğlu/Yener Ünver, İçel Suç Teorisi, 3. Baskı, Beta Yayıncılık, İstanbul, 2004.

Jakobs, Günther, Strafrecht Allgemeiner Teil: Die Grundlagen und die Zurechnungslehre Lehrbuch, 2. Auflage, Walter de Gruyter, Berlin - New York, 1991.

Jerouscheck, Günter/Ralf Kölbel, “Zur Bedeutung des so genannten Koinzidenzprinzips im Strafrecht”, Juristische Schulung (JuS), 2001.

Jescheck Hans, Heinrich/Thomas Weigend, Lehrbuch des Allgemeiner Teil, 5.

Auflage, Duncker & Humbold, Berlin, 1996.

Koca, Mahmut/İlhan Üzülmez, Türk Ceza Hukuku Genel Hükümler, 13. Baskı, Seçkin Yayıncılık, Ankara, 2020.

Köhler, Michael, “Der Begriff der Zurechnung”, Festschrift für Hans Joachim Hirsch zum 70. Geburtstag, (Hrsg.) Thomas Weigend/Georg Küpper, Walter de Gruyter, Berlin, 1999.

Kühl, Kristian, Strafrecht Allgemenier Teil, 8. Auflage, Verlag Franz Vahlen, München, 2017.

Maiwald, Manfred, “Der ‘Dolus Generalis’ Ein Beitrung zur Lehre von der Zeruchnung”, Zeitschrift für die gesamte Strafrechtswissenschaft (ZStW), Vol. 78, Issue 1-2, 1966.

Maurach, Reinhart, Deutsches Strafrecht, Allgemeiner Teil, 4. Auflage, Karlsruhe, Müller, 1971.

Maurach, Reinhart /Heinz Zipf, Strafrecht Allgemeiner Teil: Teilband 1 Grundlehren des Strafrechts und Aufbau der Straftat, 2. Auflage, C.F. Müller, Heidelberg, 1992.

Mayer, Hellmut, “Das Problem des sogenannten dolus generalis”, Juristen Zeitung (JZ), Vol. 11, No. 4, 1956.

Oğlakçıoğlu, Mustafa Temmuz, “Gübre Çukuru Olayının (BGHST 14, 193) 50 Yılı ‘Daha Da Eski Testideki Eski Şarap’”, Dokuz Eylül Üniversitesi Hukuk Fakültesi Dergisi, Prof. Dr. Durmuş Tezcan’a Armağan, C. 21, Özel S., 2019.

Oğlakçıoğlu, Mustafa Temmuz, “50 Jahre Jauchegrubenfall (BGHSt 14, 193) – Alter Wein in noch älteren Schläuchen?” Juristische Rundschau (JR), Vol. 2011, No. 3, 2011.

Otto, Harro, Grundkurs Strafrecht Die einzelnen Delikte, 7. Auflage, De Gruyter Recht, Berlin, 2005.

Ozansü, Mehmet Cemil, Ceza Hukukunda Kasttan Doğan Sübjektif Sorumluluk, Seçkin Yayıncılık, Ankara, 2007.

Özbek, Veli Özer /Koray Doğan/Pınar Bacaksız, Ceza Hukuku Genel Hükümler, 13. Baskı, Seçkin Yayıncılık, Ankara, 2022.

Önder, Ayhan, Ceza Hukuku Genel Hükümler, C. II-III, 2. Baskı, Beta Yayıncılık, İstanbul, 1992.

Özen, Mustafa, Ceza Hukuku Genel Hükümler, Adalet Yayınevi, Ankara, 2017.

Öztürk, Bahri/Mustafa Ruhan Erdem, Uygulamalı Ceza Hukuku ve Güvenlik Tedbirleri Hukuku, 21. Baskı, Seçkin Yayıncılık, Ankara, 2021.

Peters, Karl, “Mehrere Schüsse bei einer Tötung”, GA – Goltdammer’s Archiv für Strafrecht, 1958.

Pfotenhauer, Carl Eduard, Der Einfluß des factischen Irrthums und der sogenannten Verirrung auf die Strafbarkeit vorsätzlich verübter Verbrechen, Gebrúder Reichenbach, Band II, Leipzig, 1839.

Puppe, Ingeborg, Nomos Kommentar (NK), Band I, (Hrsg.) Urs Kindhäuser / Ulfrid Neumann / Hans-Ullrich Paeffgen / Frank Saliger, Nomos Verlag, Baden – Baden, 2023.

Puppe, Ingeborg, Strafrecht Allgemeiner Teil im Spiegel der Rechtsprechung 4.

Auflage, Nomos, Baden-Baden, 2019.

Puppe, Ingeborg, Vorsatz und Zurechnung, Heidelberger Forum, Decker & Müller, Heidelberg, 1992.

Roxin, Claus/Luís Greco, Strafrecht Allgemeiner Teil, Band I, 5. Auflage, C.H.Beck, München, 2020.

Roxin, Claus, “Zur Erfolgszurechnung bei vorzeitig ausgelöstem Kausalverlauf”, Goltdammer’s Archiv für Strafrecht (GA), 2003.

Roxin, Claus, Gedanken zum “Dolus Generalis”, Kultur, Kriminalität, Strafrecht:

Festschrift für Thomas Würtenberger zum 70. Geburtstag, Duncker & Humblot, Berlin, 1977.

Roxin, Claus, “Gedanken zur Problematik der Zurechnung im Strafrecht”, Festschrift für Richard M. Honig zum 80. Geburtstag, O. Schwartz, Göttingen, 1970.

Rudolphi, Hans Joachim, Systematischer Kommentar zum Strafgesetzbuch (SK - StGB), Band I, (Hrsg.) Jürgen Wolter, 9. Auflage, Carl Heymanns Verlag, Köln, 2016.

Sancinetti, Marcelo A., “Dolus Generalis” und “strafrechliches Glück”, Festcshrift für Claus Roxin zum 70. Geburtsatg, (Hrsg) Bernd Schünemann / Hans Achenbach / Wilfried Bottke / Bernhard Haffke / Hans-Joachim Rudolphi, Walter de Gruyter, Berlin / New York, 2001.

Sanz Morán, Ángel José, “Aproximación al problema del denominado ‘dolus generalis’”, InDret: Revista para el Análiss del Derecho, No. 3, 2016.

Sanz Morán, Ángel José, El Concurso de Delitos: Aspectos de Política Legislativa, Ediciones Universidad de Valladolid, Valladolid, 1986.

Schmidhaüser, Eberhard, Strafrecht Allgemeiner Teil, 2. Auflage, J.C.B. Mohr, Tübingen, 1975.

Schroeder, Friedrich-Christian, Leibziger Kommentar StGB, 11. Auflage, Walter de Gruyter, Berlin/New York, 1994.

Sowada, Christoph, “Der umgekehrte ‘dolus generalis’: Die vorzeitige Erfolgsherbeiführung als Problem der subjektiven Zurechnung”, Juristische Ausbildung (JURA), Vol. 26, Issue 12, 2004, s. 815.

Stratenwerth, Günther/Lothar Kuhlen, Strafrecht Allgemeiner Teil: Die Straftat, 6.

Auflage, Verlag Franz Mahlen, München, 2011.

Sternberg-Lieben, Detlev /Frank Schuster, “Vorsätzliches und fahrlässiges Handeln”, Schönke/Schröder Strafgesetzbuch Kommentar, 30. Auflage, C.H.Beck, München, 2019.

Vogel, Joachim/Jens Bülte, “Irrtum über Tatumstände”, Strafgesetzbuch Leipziger Kommentar, 13. Auflage, Walter de Gruyter, Berlin, 2020.

von Buri, Maximilian, “Zur Lehre von der Tödtung”, Goltdammer's Archiv für Strafrecht (GA), Band 11, 1863.

von Hippel, Robert, Lehrbuch des Strafrechts, Springer Verlag, Berlin Heidelberg, 1932. von Liszt, Franz/Eberhard Schmidt, Lehrbuch des Deutschen Strafrechts: Band I Einleitung und Allgemeiner Teil, 26. Auflage, Walter de Gruyter, Berlin Leipzig, 1932.

von Rengier, Rudolf, Strafrecht Allgemeiner Teil, 11. Auflage, C.H. Beck, München, 2019.

von Weber, Heinrich Benedikt, “Ueber die verschiedenen Arten des Dolus”, Neues Archiv des Criminalrechts, No. 7, 1825.

Welzel, Hans, Das Deutsche Strafrecht, 11. Auflage, Walter de Gruyter, Berlin, 1969.

Wessels, Johannes/Werner Beulke/Helmut Satzger, Strafrecht Allgemeiner Teil;

Die Straftat und ihr Aufbau, 47. Auflage, C.F. Müller, Heidelberg, 2017.

Walter, Tonio, “Irrtümer auf Tatbestandsebene”, Handbuch des Strafrechts, Band 2: Strafrecht Allgemeiner Teil, (Hrsg.) Hilgendorf/Kühl/Valerius, C.F. Müller, Heidelberg, 2019.

Wolter, Jürgen, “Zum umgekehrten dolus generalis. Subjektive Zurechnung eines beendeten (tauglichen) Versuchs bei objektiver Zurechnung der Vollendung”, Goltdammer’s Archiv für Strafrecht (GA), 2006.

Yüce, Turhan Tufan, Ceza Hukuku Dersleri, C. I, Şafak Basım ve Yayınevi, Manisa, 1982.