Attempts and Trying

The Philosophical Problem of Attempt

The philosophy of action investigates what agents do, how they do it, and what makes an occurrence intelligible as their doing. Attempts expose an important pressure point in that investigation. Suppose a person tries to open a door that has been locked from the other side. The door remains shut. Nevertheless, the person may have reached for the handle, turned it, pulled against the resistance, reconsidered the grip, and made several further movements under a continuing intention to open it. An outcome description says that the door was not opened. An action description says that the agent tried to open it. The second description identifies a real practical episode whose intelligibility survives failure.

The case raises several questions at once. What must occur before the assertion that someone tried becomes true? Does an intention by itself amount to trying? Can trying exist without observable bodily motion? Is there always some identifiable deed constituting the attempt? Does trying name an event, an activity, a relation to an end, or a conditional fact about what an agent would have done under favorable circumstances? How should the analysis distinguish failing to do something from never attempting it? And why are agents sometimes responsible for attempts even when the projected outcome is absent?

These questions matter because much philosophical vocabulary is organized around accomplished actions. We describe someone as crossing a street, delivering a letter, breaking a window, rescuing a child, or finishing an argument. Such descriptions invite a perfective reading on which the named transition is complete. Attempts disclose the practical structure preceding completion. One can undertake the crossing, initiate delivery, direct force toward the window, intervene in a rescue, or construct a proof while remaining short of the described terminus. The ontology of action must account for this incompleteness without erasing the actions that actually occur.

There is also a problem of individuation. When the person turns the locked handle, are there two events, a turning and a trying to open, or one event describable in two ways? If the person tries again, are there two attempts of the same kind? If a sustained effort comprises many adjustments, are these independent attempts or phases of one larger undertaking? An account that counts only physical movements risks losing the directed unity of the episode. An account that counts only intentions risks classifying inactive wishing as genuine trying.

The principal thesis developed here is that attempt is a practical classification of an agent's situated engagement with an end. Its truth depends on more than intended success and less than achieved success. The category therefore brings intention, initiation, world-directed means, and practical control into a common explanatory structure. It also identifies an irreducible feature of agency: agents can conduct themselves toward states of affairs whose realization remains open.

Definition and Analytical Scope

An attempt to bring about an end E is an episode in which an agent, guided by an operative practical orientation toward E, initiates or sustains conduct that is intelligibly organized as a way of realizing E, under circumstances in which realizing E is not guaranteed by undertaking that conduct. This working definition identifies a phenomenon rather than imposing a new canonical theory. Each clause responds to a recurrent philosophical difficulty.

The operative orientation may be an occurrent intention, a settled plan presently being executed, or a purposive practical commitment whose content can be identified in the agent's performance. Someone may attempt to stop a rolling object with a quick movement without forming a verbally articulated plan. What matters is that the movement belongs to an intelligible practical course toward stopping it. The orientation distinguishes a purposive interception from the accidental collision of a body with the object.

Initiation or sustained conduct supplies the difference between merely considering an act and attempting it. Thinking about sending a message tomorrow, resolving to send it, preparing a draft, and pressing the send command represent different positions along a practical sequence. Whether drafting already counts as attempting delivery depends on the unit of action under examination and the relevant causal route. A distant preparatory step may be part of a project without yet constituting an attempt at the immediately described result. There is no universal physical threshold separating every preparation from every attempt; the threshold depends on how the projected action is individuated.

Intelligible organization as a means supplies a rational and causal orientation without demanding that the means be competent. An agent may try to start a machine by pressing an ineffective switch because the agent mistakenly believes it to be the starter. The attempt is real although the instrumental belief is false. Yet if the putative conduct is wholly disconnected from the end and the agent supplies no coherent understanding of how it could count as attempting, the attribution becomes unstable. Errors can coexist with agency, while arbitrarily redescribing unrelated movements as attempts empties the category of meaning.

Contingency of success distinguishes attempting from succeeding. Someone who tries and succeeds ordinarily performs a successful attempt, although ordinary speech sometimes reserves the verb 'try' for occasions of uncertainty, caution, or difficulty. Someone who tries and fails performs an unsuccessful attempt. These are assessments of the same broader practical form. Not every action is linguistically reported as an attempt, and philosophers disagree whether every intentional action ontologically includes trying. This monograph separates the metaphysical question from the pragmatic uses of the word.

The subject also has boundaries. General theories of agency examine what qualifies something as an agent; theories of intention examine a practical attitude; theories of basic action investigate the stopping points of action explanation; theories of responsibility investigate normative attribution. Attempts and Trying studies the relations among these objects at the point where directed conduct is undertaken without an assurance of completion. Its distinctive research questions concern commencement, failure, persistence, fulfillment, and the status of unsuccessful practical engagement.

Aristotle on Voluntariness Choice and Achievement

In Book III of the Nicomachean Ethics, Aristotle distinguishes voluntary from involuntary conduct and analyzes choice, prohairesis, as central to ethical agency. The voluntary is connected with the originating principle in the agent and with knowledge of the relevant particulars. Choice involves deliberation about matters within the sphere of action. These distinctions supply an early framework for understanding why practical engagement cannot be reduced to the state of affairs eventually produced. A person may act voluntarily yet fail to obtain the result sought, and the explanation of that conduct still depends on the agent's practical orientation (Aristotle, trans. Ross, Book III, chapters 1–5; primary text).

Aristotle's distinction between wishing and choosing is especially useful. One may wish for a distant or impossible good, but deliberation and choice concern the manner in which something can be pursued through one's activity. The concept of an attempt occupies this practical territory: the agent takes up an available course of action in relation to a desired end. A person may wish for health without attempting a cure; another may choose and undertake treatment despite its uncertain efficacy. Undertaking introduces a new practical relation to the end.

The Aristotelian view also establishes that ethical evaluation attends to information and circumstances. The same bodily episode may be voluntary or involuntary depending on what its agent knows, and a mistaken judgment about means may alter assessments of deliberative quality without making the conduct an accidental occurrence. These distinctions anticipate an important division between whether an episode counts as trying and whether it counts as rational, skillful, or responsible trying. The actuality of an attempt and its quality are different questions.

Aristotle does not present the contemporary ontology of trying as a separate research program. It would be historically misleading to attribute modern discussions of disembodied trying, causal deviance, or counterfactual attempt conditions to his texts. His contribution is conceptual: action analysis requires attention to the agent's initiating principle, objects of deliberation, and circumstances of performance. His distinction between ends and what conduces to ends helps explain how efforts toward an end can be genuine even when the end is unrealized.

A further lesson follows from the relation between activity and completion. An ethical action can be assessed in its ongoing performance. Courage, moderation, and practical judgment concern what one does in circumstances requiring action, not exclusively what the world later permits one to achieve. For the theory of attempts, this means that achievement cannot be the sole ground of practical description or evaluation.

Will Exertion and the Early Modern Inheritance

Early modern debates about action develop several related questions: how the will directs bodily movement, what distinguishes voluntary motion from reflex or compulsion, and how an agent can be active when bodily powers are obstructed. Descartes's separation between thought and extended matter leaves a distinctive problem of interaction: intending a bodily motion and producing that motion belong to differently described domains. In The Passions of the Soul, the dependence of voluntary movement on bodily mechanisms makes room for a divergence between willing and the resulting mechanical process (Descartes, 1649/1989, arts. 17–18 and 41–44).

Gilbert Ryle challenges the explanatory use of separately postulated volitions in The Concept of Mind: describing a movement as voluntary by invoking another inner act can relocate, rather than resolve, the problem of what makes the performance agentive (Ryle, 1949/2003). Locke distinguishes willing from desiring and connects voluntary actions with the mind's power to begin or forbear bodily movements. In An Essay Concerning Human Understanding, Book II, chapter 21, the discussion of power, volition, and uneasiness shows why a desire for an outcome is not sufficient to characterize a particular bodily event as one's attempt to secure it. A person may desire a result without choosing to move toward it, and may choose a movement without possessing the ability to bring about the intended further outcome (Locke, 1690, II.xxi; textual edition).

The historical category of volition was often asked to perform two tasks. It was supposed to explain how an action begins and to establish the agent's distinctive contribution to what happens. These tasks become difficult when the explanation locates an additional inner act behind each bodily act. If moving one's arm requires a prior act of willing the movement, one may ask what makes this willing itself active. Infinite regress threatens if every action must be caused by another action of the same explanatory kind. A theory of volition therefore needs an account of initiation that does not merely duplicate what it purports to explain.

The problem is sharpened by cases of impairment. An individual may direct an effort toward moving a limb while paralysis prevents movement. Ordinary descriptions sometimes treat this episode as an effort to move; others reserve action for successful motor execution. Both are supported by something recognizable: there may be genuine first-person exertion, yet the target bodily transition does not occur. The case forces a distinction between the exercise of a capacity and the realization of its characteristic effect.

The early modern inheritance contributes a map of the problem rather than a single settled solution. Volition, desire, bodily causation, power, and performance can come apart. Modern philosophy of attempts develops by asking how many of these are required for an agent genuinely to try, and how they can be related without treating success as necessary.

Wittgenstein on Raising an Arm and Trying

Wittgenstein's Philosophical Investigations returns to the apparent contrast between an arm's rising and one's raising it. The famous question in section 621 asks what remains if the fact that an arm rises is subtracted from the fact that one raises it. The temptation is to posit a distinct interior act or contribution that, when added to a bodily occurrence, turns it into an action. The surrounding remarks about willing and trying destabilize the picture of action as a movement plus a privately observed mental accompaniment (Wittgenstein, 1953, §§614–628; see edition record).

Wittgenstein notices that attempts and acts are not always reported with the same grammar. When one raises one's arm ordinarily, one need not separately undertake a preliminary effort called trying to raise it. Yet when there is resistance or inability, trying becomes salient. This grammatical difference reflects the settings in which an effort must be identified, but it does not settle whether every intentional bodily performance shares a common metaphysical constituent named trying.

There are at least three interpretations of the remarks. The first treats them as a warning against identifying an action by supplementing a neutral motion with a private feeling. The second emphasizes the context dependence of verbs of trying: what counts as a genuine effort depends on a practice of identifying tasks, capacities, and standards of performance. The third reads Wittgenstein as pressing a general question about the intelligibility of volition as a causal intermediary. These readings can overlap, and none supports the simple conclusion that bodily action is an entirely mechanical occurrence.

Consider a person who deliberately lifts an arm to vote. The action is intelligible through the voting practice, the agent's reasons, and the directed movement. An additional interior episode is not needed to explain the difference between this case and an involuntary muscular contraction. Conversely, a person instructed to raise a paralyzed arm may genuinely exert toward the motion even though the limb remains still. In that case the term 'trying' may identify an effort whose evidence differs from the normal evidence for successful raising.

The philosophical lesson is methodological. We should not derive a hidden ontology of acts from a simple subtraction metaphor. Nor should we collapse attempted movement into successful movement because the same ordinary verb sometimes covers both. Inquiry must determine the practical and causal circumstances supporting attributions of attempting. Wittgenstein makes the apparently familiar relation between doing and trying a site of philosophical investigation.

Anscombe on Intention and Practical Description

In Intention, G. E. M. Anscombe investigates intentional action through the question of why an agent is doing something. Her account examines the descriptions under which an action is intentional, the way an agent's answer places movements within a practical order, and the distinctive knowledge accompanying intentional activity (Anscombe, 1957/1963, particularly §§5–6, 23–32 and 45; bibliographic record).

The descriptive character of action is essential for attempts. A person may intentionally press a button, attempt to activate an alarm, and inadvertently disable another device. The physical contact may be one event or part of a continuous episode, but the agent's action is intentional under some descriptions and unintended under others. Likewise, a failed attempt may contain intentional subactions without containing the intended terminal act. The agent intentionally presses the button; the alarm never activates. The truth of the first description grounds the attempt to activate it.

John Searle's later account of intention in action distinguishes prior intention from the intentional organization of performance and analyzes satisfaction conditions linking intended activity to the world (Searle, 1983, chapter 3). These conditions are particularly revealing where the attempt's causal route is activated but its target is unrealized.

Anscombe's emphasis on practical knowledge raises a difficulty. If agents characteristically know what they are intentionally doing, how can they know they are opening a door when the door remains locked? The solution begins by distinguishing an ongoing practical description from a completed achievement. An agent may know, without observation, that they are trying to open the door, using the handle to open it, or applying pressure toward opening it. The claim that the door has actually opened depends on whether the relevant world-directed conditions have been satisfied.

This distinction is not a license to let the agent's self-description determine every fact. Practical knowledge is answerable to what the agent can actually perform. Someone may mistake the object grasped, fail to make contact, or misunderstand the mechanism. An account of attempts must therefore explain how first-person direction and external correction jointly determine the content of action descriptions.

Anscombe also helps clarify means-end unity. The agent's conduct is not merely a collection of independently motivated events. Turning a handle may be part of opening a door; opening a door may be part of entering a building; entering may be part of delivering assistance. When the larger action fails, the lower-level intentional actions can remain. A theory of trying must follow these nesting relations and identify the point at which the intended end was pursued.

The resulting insight is that trying is always trying under some practically relevant description. The same episode can be a successful attempt to turn a handle and an unsuccessful attempt to open a door. There is no contradiction, since success conditions differ with the ends under which the episode is classified.

Reasons Causes and the Production of Action

Donald Davidson's “Actions, Reasons, and Causes” develops a causal approach to action explanation. A primary reason combines an attitude favorable toward actions of a certain kind with a belief connecting the action to a desired feature or end. Such reasons both rationalize and, in the relevant circumstances, cause actions (Davidson, 1963, pp. 685–700; original article). This framework can explain why attempts are more than desired outcomes. An agent's reason must be implicated in conduct actually undertaken.

Take a person who wants to summon help and believes that pressing a button will sound an alarm. When the person presses the button for this reason, the occurrence is explicable as an attempt to summon help even if the alarm is defective. The agent's intention and the relevant bodily activity form an intelligible causal sequence. The failure of the alarm does not retroactively erase the intentional pressing.

The causal account, however, encounters the problem of deviant causal chains. An agent might become so anxious about having to speak publicly that the anxiety causes a convulsion activating the alarm. An intention to activate it exists and contributes causally to its activation, yet the activation need not count as the agent's intentional attempt. The causal route must sustain action-guiding agency rather than merely lead from a reason to an outcome.

Harry Frankfurt proposes that action involves a distinctive sort of guidance and control during its occurrence. The agent's relevant contribution is evident in the ongoing regulation of what happens, rather than exclusively in a preceding mental cause (Frankfurt, 1978, pp. 157–162; article). This approach is particularly attractive for attempts because their paths commonly include correction, monitoring, and response to interference. An effort can persist through fluctuations that frustrate immediate progress toward the end.

J. David Velleman likewise questions whether a bare causal relation between an agent's motivating attitude and an event adequately preserves the agent's role in what happens (Velleman, 1992). This pressure reinforces the need to articulate the guidance relation rather than treating any causation by reasons as action.

There is a tension between a wholly causal and a purely rationalizing analysis. The former risks admitting deviant sequences; the latter risks admitting mere intentions unconnected to performance. A satisfactory account must explain how the agent's reasons are operative in a course of activity. It is not necessary to deny causation; the point is to identify the mode of causation appropriate to agency.

Attempts also demonstrate that explanatory success is different from causal success. A reason can successfully explain why someone moved a lever although the machine did not start. The attempt's practical intelligibility survives the failure of the mechanism that connects the lever to the desired outcome. This makes attempted action a useful test case for any general theory of intentional conduct.

Volition and the Ontology of Trying

Volitional theories take seriously the possibility that trying to perform an act is itself an exercise of will. Brian O'Shaughnessy's “Trying” made this issue central to the modern philosophy of action, prompting questions about what trying adds to intending and doing (O'Shaughnessy, 1973, pp. 365–386; article record). Hugh McCann subsequently developed volitional interpretations of basic action and paralysis, asking whether willing or trying remains an act when ordinary motor consequences are absent (McCann, 1974; 1975).

The attraction is clear. If an individual tries to raise a paralyzed arm, one may be able to identify a direction of activity that persists despite the absence of movement. A theory that recognizes trying as a basic exercise of agency can explain what the individual is doing. A view that treats all action as manifest bodily movement has a harder time accounting for such episodes unless it identifies some other bodily performance, such as muscular activation or neural command, that constitutes the attempt.

Jennifer Hornsby's analysis of trying draws attention to the distinction between acting and trying to act and to the contexts in which these descriptions become philosophically salient (Hornsby, 2010). It suggests that a theory needs semantic discrimination as well as an ontology of constituent events.

The ontology faces a different challenge. If trying is an additional event, how is it distinguished from the intention causing it, the motor command initiating it, and the bodily movement sometimes following it? An isolated inner event cannot be postulated simply because a language of trying seems appropriate. The theoretical cost of multiplying events must be justified by explanatory work: accounting for failed bodily execution, specifying first-person initiation, and distinguishing attempted action from passive happenings.

Frederick Adams and Alfred Mele examine the intention–volition debate, exploring whether intentions, tryings, and feedback can perform functions traditionally assigned to volitions (Adams & Mele, 1992). Their approach recommends distinguishing functional components before assuming that every component corresponds to an ontologically separate act. An intention can settle what is to be done; a trying episode can connect that settling to performance; sensory and other feedback can influence continuation. The components may be functionally separable even where the underlying physical process is continuous.

A rival account regards trying as the agent's relevant conduct under a purposive description. On that view, pressing a switch, straining against resistance, or issuing a command can be the attempt without any further trying event. Such an account avoids duplication but must say how trying is present when overt conduct is blocked. The paralysis cases therefore remain an especially important test.

The most defensible methodological starting point is ontological restraint combined with descriptive seriousness. An attribution of trying identifies something real about the practical episode. Whether that reality requires a separately countable event must be argued, not inferred directly from the availability of a verb.

Intention to Act and Intention to Try

Intending to do something and intending to try to do it are related but distinguishable. A person may intend to complete a task and act accordingly; another may undertake a difficult test while anticipating failure and intending, above all, to make an attempt. Alfred Mele's “She Intends to Try” examines the relation between trying, intending, and the intentional pursuit of an outcome, challenging overly strong equations among them (Mele, 1989, pp. 101–106; article).

The distinction matters because there are varying attitudes toward success. An actor may make a serious attempt to solve an unfamiliar problem while having little confidence in solving it. The actor can intend to try and perhaps hope to succeed without settling in advance that success will occur. Conversely, an agent may intend to reach a destination and routinely undertake the necessary steps without forming the separate higher-order intention to try to reach it.

There is also a logical difference between planning a performance and performing it. One can decide now to attempt a rescue tomorrow. This decision is an intention concerning a future attempt, not yet the attempt itself. The transition to trying requires practical uptake: initiating the relevant response or placing oneself in an operative process directed to the end. Mere certainty about what one will attempt tomorrow is insufficient.

Bratman's planning theory illuminates the temporal relation between intention and implementation. Plans coordinate action over time, guide later choices, and establish constraints for further deliberation (Bratman, 1987). A plan to repair an instrument may persist throughout a series of unsuccessful trials. Some stages may be attempts to remove a component, while the entire sequence may be a continuing attempt at repair. The plan provides practical organization without guaranteeing the status of any particular step.

The phenomenon also complicates desire-based definitions. A person may attempt a task reluctantly because of duty, and may undertake a test without wanting the tested result to occur. Mele's “He Wants to Try” explores how the motivation to make an attempt may differ from the motivation to obtain an outcome (Mele, 1990). The target of trying cannot always be read directly from the agent's strongest desire.

A precise account should therefore avoid inferring from “tried to do E” that the agent held one uniquely specified higher-order intention. What is necessary is an operative orientation giving the course of conduct E-directed significance. That orientation may include an intention to achieve E, a plan to test whether E can be achieved, or a commitment to make a sincere attempt where a situation demands one.

The Success Conditions of an Attempt

Success is indexed to an intended end under an action description. The success condition for attempting to open a door is that the relevant door opens through the agent's appropriate activity; for attempting to discover a theorem, the condition involves reaching a warranted result through an appropriate inquiry; for attempting to persuade someone, it concerns a change of mind attributable to the persuasive performance. These are distinct conditions, and their differences cannot be eliminated by describing all attempts as generic effort.

An important distinction separates success at trying from success at the underlying task. To say that someone successfully tried to lift a weight can sometimes mean that the person made a sincere and competent attempt, even if the weight remained immobile. In a different context the phrase means that the person lifted the weight. The ambiguity reflects two evaluation levels: whether the attempt was genuinely undertaken and how it was performed, and whether its target result occurred.

Suppose a scientist attempts to reproduce an experimental effect. The protocol is executed competently and the effect does not appear. The attempt may be scientifically valuable, methodologically sound, and unsuccessful relative to the specified target. Calling it a success without qualification would obscure its outcome; calling it an absence of action would erase the protocol's implementation. A theory of attempts needs both evaluative dimensions.

John Hyman's examination of action, knowledge, and will underscores the need to relate powers and rational capacities to the acts agents actually perform (Hyman, 2015). Whether an achievement is attributable to an agent turns on the exercise of the appropriate power in the episode, not merely on an outcome description.

Another complication is overachievement or accidental success. If an intended outcome occurs through an unrelated coincidence while the agent is trying to produce it, the attempt has not necessarily succeeded in the relevant agentive sense. Someone trying to unlock a door presses a defective code, but another person opens it from within at exactly that moment. The door opens, yet the agent has not unlocked it. A minimally adequate success condition therefore requires appropriate dependence between the agent's conduct and the achieved outcome.

The same point applies to ordinary cases of partial success. One can make progress toward a goal, complete several subgoals, and still fail at the final objective. A rescue attempt may succeed at establishing contact yet fail to bring the person to safety. Each layer has its own success criterion. Explanations should preserve the nested structure instead of issuing one undifferentiated verdict.

The philosophical consequence is that attempts are not defined by the absence of success. Success and failure are outcomes or assessments of attempts. The category of attempting must be capable of encompassing both, while preserving the conditions that distinguish successful achievement from coincidental satisfaction of the intended state of affairs.

Failed Attempts and the Reality of Agency

A failed attempt is a real practical episode directed toward an end whose completion condition is not met. The concept is positive: it identifies undertaking, guidance, and world-directed activity. Its failure predicate identifies an unrealized achievement. This analysis avoids treating an unsuccessful attempt as a missing action and avoids treating it as a complete achievement under another name.

Consider a person who attempts to post a letter by placing it into a collection box that has been removed from postal service. The person may have written the address, applied postage, traveled to the box, and deposited the envelope. If the intended act is getting the letter into the postal network, the result fails. Yet multiple intentional actions remain. The attempt is partly constituted by those actions as directed toward the wider end.

The case teaches that action descriptions are layered. “Depositing an envelope” may describe something done, while “sending the letter” describes something unsuccessfully pursued. The distinction is not arbitrary because the action descriptions have different completion conditions and different causal dependencies. One must identify which practical end the agent undertook and what action-guided conduct took place.

A failed attempt can also occur at the most immediate bodily level. A person tries to move a limb and no movement follows. Here it may be inappropriate to redescribe the effort as an intentional movement that simply happened to be unsuccessful. The question becomes what, if anything, is performed in the effort itself. Philosophers disagree whether such cases require a volitional event, an embodied motor initiation, or a conditional account of what the agent would do under unimpeded circumstances. The disagreement concerns the ontology of the episode, not simply whether the person's practical orientation is intelligible.

Another objection states that a failed attempt can be acknowledged only if there was some successful action of trying. This assertion is potentially informative, but it risks turning “trying” into a formal success condition by definition. Saying that the attempt took place explains why the agent can be credited or blamed for the undertaking; it does not independently establish what the agent did to initiate that undertaking. For that, one must analyze specific capacities, movements, decisions, commands, or circumstances.

Failures illuminate agency because they separate direction from realization. In a deterministic or a probabilistic world, the external conditions of success may differ from what the agent expects. Philosophical explanations still need to locate what belongs to the agent's activity. A genuine attempt is therefore evidence of a practical history even where the anticipated future never occurs.

Basic Action Paralysis and Bodily Execution

The theory of basic action asks whether some things agents do are done directly, rather than by doing another thing as a means. Raising an arm may be basic in an ordinary context, while opening a door is commonly achieved by moving one's hand and the handle. An attempt at a nonbasic end can be constituted by the successful performance of more basic actions, even where the remote end fails.

Paralysis unsettles this model. A subject who tries to raise an arm may experience initiation, concentration, or exertion without any visible raising. If moving a limb is the most basic bodily act in the case, there seems to be no successful lower-level action to constitute the attempt. McCann's analyses of volition and paralysis explore this terrain, challenging the assumption that the outward movement must occur for any agentive episode to be present (McCann, 1974; 1975).

One possible response is that the agent performs a neural or muscular act that ordinarily produces movement. This relocates basic action to a finer physiological level. It must nevertheless explain how an agent can intentionally perform a neurological event under a description of which the agent ordinarily knows nothing. Agents control bodies through learned skills and practical capacities; they do not typically direct individual neuronal firings as explicit projects.

A second response identifies a volitional exercise. Its virtue is the ability to distinguish active trying from passive anticipation of motion. Its problem is explaining the relation between the supposed basic volition and an otherwise inactive body. The more the theory detaches volition from embodied causal organization, the more difficult it becomes to distinguish a genuine motor attempt from intense wishing.

A third response adopts a counterfactual or capacity-based analysis. An attempt can be true because the agent is engaged in an appropriately organized process that would have issued in performance absent disabling conditions. This approach preserves the relevance of the body's normal action system while allowing an obstacle to prevent observable output. It must handle cases in which the disabling condition also disrupts the process in question.

These accounts need not be judged only by their treatment of extraordinary cases. Ordinary bodily action also involves control hierarchies, feedback, learned abilities, and failures of execution. The relevance of paralysis lies in its power to separate layers that usual success conceals. A monograph on trying must therefore examine bodily implementation as a substantive philosophical issue rather than assume that the intention–movement connection is transparent.

Control Guidance and Causal Deviance

A theory of trying must establish the relation between the agent and the unfolding practical process. Control is central because an external event can satisfy a person's objective without being the person's successful action. Likewise, causal effects of intention can occur without being the agent's controlled attempts. Frankfurt's emphasis on guidance helps identify what differentiates action from sequences that merely originate in mental states (Frankfurt, 1978).

Control admits degrees and forms. An agent may exercise fine motor control while attempting to sign a name, supervisory control while programming a machine, or deliberative control while directing a collaborative project. The distribution of bodily motion, sensory feedback, and delegated operations differs, but each can support an attribution of trying when the relevant process is organized under the agent's practical direction.

The deviant-causation objection concerns a sequence in which a reason causes an outcome through a path not appropriately governed by agency. Suppose a person intends to catch a ball, becomes startled by the prospect, and reflexively jerks a hand into the ball's path. The intention may be causally implicated in the catching, but the reflex alone does not establish that the person intentionally attempted the catch through that motion. To establish an attempt, we require evidence of performance-guiding engagement.

Yet control cannot be made too demanding. Many genuine attempts unfold in uncertain environments, using incomplete information and instruments that respond unpredictably. A person who pulls a rope to free a trapped object is attempting a release even when the resistance prevents effective control of the object's motion. The agent controls the undertaking to some extent while lacking control over the target's result.

A useful distinction is therefore between local control of the attempt's constituting conduct and global control of its outcome. The first identifies what the agent can initiate or regulate in the circumstances; the second concerns whether the intended world-state can be produced. A person may have the first without the second. If an account demands global control as a condition of attempting, it excludes most uncertain undertakings, including many paradigmatic ones.

The analysis leads to a graded notion of responsible performance. Agents may be held accountable for how they conduct an attempt, how they respond to emerging information, and whether they withdraw when appropriate. Attribution can remain meaningful despite limited outcome control. This possibility is foundational for subsequent ethical assessment.

Effort Skill and the Phenomenology of Trying

Trying is often associated with felt effort. The association is strongest where circumstances resist the agent: lifting a heavy object, focusing on a difficult calculation, recovering a lost word, or making a movement under impairment. However, effort and attempt do not have identical extensions. Skilled action often involves a stable, unstrained flow, and a person can deliberately attempt a routine task with little noticeable exertion. Conversely, bodily exertion can occur during an activity with no determinate intended external achievement.

The phenomenology of agency concerns how acting appears to the agent. Elisabeth Pacherie develops a framework for understanding the experience of action, including different levels of agency, intentional organization, and the sense of control (Pacherie, 2008; article). This research is helpful for explaining why attempts are sometimes experienced as failures of execution even before an observer determines their outcome.

The sense of effort may arise from anticipating or responding to resistance. It can signal that an intended action is difficult, that feedback is discordant, or that control must be intensified. Yet felt effort is not an infallible marker of either agency or successful causal engagement. A subject may feel intense effort during an ineffective movement or may execute a controlled skilled action with little reflective feeling.

A philosophical account should accordingly distinguish three questions. What causes the experience of effort? What practical activity is the agent undertaking? And what evidence justifies an observer or the agent in attributing a genuine attempt? The answers may overlap without coinciding. An abnormal feeling may accompany successful action; an ordinary feeling may accompany action failure; an attempt may occur without an explicit experience of trying.

Skill introduces an additional complication. Expert performers continuously adjust means in response to relevant features of a task. Their engagement can be intentional without requiring a separate conscious command at every movement. An analysis that requires deliberate mental effort for every attempt would imply that the most fluid performance is less agentive than the least competent struggle. That result confuses phenomenal strain with practical organization.

The stronger approach treats effort as one possible mode of trying and one form of evidence about it. Trying is fundamentally described through practical orientation and undertaken conduct. Phenomenology clarifies how agency is lived, while external action analysis determines how the episode is structured and what it accomplishes.

Trying Across Time

Attempts occupy time. A brief effort to push a button may be nearly instantaneous; an attempt to discover a mathematical proof can continue for months; an attempt to repair a relationship can involve intermittent actions across years. A theory that counts only momentary volitions cannot easily explain the persistence and unity of extended undertakings. Nor can a theory that counts every physical operation as an independent attempt explain how projects remain numerically and practically coherent.

Michael Bratman's account of plans provides resources for understanding temporally extended commitment. Plans coordinate earlier and later activities, settle some questions while leaving others open, and permit adjustment as information changes (Bratman, 1987). An agent attempting a complex task typically operates through such structured commitments. A failed step may prompt revision while the overall attempt continues.

The distinction between accomplishment and process is philosophically important. A person can be crossing a street before having crossed it. This progressive description is true during a partially completed event, while the perfective completion claim can remain false. Michael Thompson's treatment of practical and processual description makes the temporal organization of action a central issue (Thompson, 2008). Attempts are often intelligible because the agent's activity is underway toward a possible completion rather than because completion has already occurred.

Persistence does not mean that an attempt remains active whenever an agent retains the relevant intention. A person may suspend a project for sleep or illness and later resume. Whether the intervening period belongs to one attempt, several attempts, or a standing undertaking depends on the criteria governing the practical project. A continuing plan can unify temporally separated episodes without making every moment an act of trying.

Likewise, repetition creates problems of individuation. A runner may attempt a jump three times. Each launch has an immediate success condition; the training session may constitute one larger attempt to achieve a performance standard. The relevant units are established by an interaction of the agent's practical organization, interruptions, resets, and the social rules of the task. Counting movements alone misses these criteria.

The temporal structure matters for responsibility and knowledge. An agent can revise a defective strategy, abandon a dangerous effort, or change an objective while continuing to act. What began as one attempt may become another as its practical end is redefined. Philosophical description must therefore be sensitive to the changing purposes and boundaries of unfolding activity.

Ruben Payton and the Metaphysics of Attempts

Recent analytic work directly addresses whether attempts require independent events or particulars. David-Hillel Ruben's “A Conditional Theory of Trying” develops a counterfactual analysis of trying, resisting the assumption that every true trying attribution identifies a distinct trying particular (Ruben, 2016, pp. 271–287; article). His later The Metaphysics of Action: Trying, Doing, Causing places the question within a broader account of action and its causal relations (Ruben, 2018; publisher).

The central attraction of a conditional approach is that it can connect unsuccessful attempts to performances that would have occurred had obstacles not intervened. Consider an agent whose motor system is oriented toward raising a limb although paralysis blocks the ordinary movement. A suitable counterfactual might specify what the agent would do in a nearby unimpeded situation. Such an approach can preserve the significance of capacity and direction without positing a new private action whenever an attempt fails.

Counterfactual explanations, however, need careful specification. If an agent is unable to act because an obstacle prevents the initiation of any practical process, should a bare willingness to act count as trying? A conditional such as “the agent would have acted if circumstances were different” is too permissive unless it identifies which differences are admissible and what actual engagement grounds the conditional. Persons often would act under alternative conditions while currently doing nothing.

Jonathan D. Payton's “Attempts” challenges conditional analyses and defends the view that genuine attempts require an occurrence, or attempt event, with a more substantial positive ontology (Payton, 2022, pp. 363–382; article). This disagreement is especially instructive because it concerns the metaphysical truthmakers of attributions: what, in the actual world, makes it true that an agent attempted something?

Payton develops this claim through a process view: an attempt is initiated and sustained by an appropriate practical orientation and may extend from neural preparation into muscular and overt activity. The process can remain entirely or largely internal when execution is blocked. In his analysis of so-called naked trying, a patient may attempt to raise an immobilized arm without performing another action as a means of raising it (Payton, 2022, section 2.1; accessible primary article). He contrasts this view with volitionism, which identifies attempts as an exclusively mental kind of event, and teleologism, which identifies trying to do something with another act done in order to achieve it. His semantic argument proceeds from the possibility of event-modifying adverbs and perceptual reports about trying to the existential commitments of trying sentences. This is a substantive challenge to Ruben's interpretation of such sentences as conditionals; its success depends on whether the semantic evidence genuinely requires a distinct class of attempt events rather than ordinary actions described under a practical aspect.

An event-based view has intuitive force. When a person tries to turn a locked key, there is real movement and exertion. When a computer operator sends a command to a remote machine that does not respond, there is real intervention. The attempt can be located in what was done, regardless of the absence of the target outcome. But if the view insists on a distinct event of trying over and above the relevant actions, the burden of ontological justification returns.

A conditional theory explains why an agent's present practical orientation can bear on an unrealized act. An event theory explains why attributions of attempting often record an actual episode of agency. The positions are not exhausted by their most radical formulations: a theory can require an actual undertaking while using counterfactuals to analyze the undertaking's connection to its end. The most productive question is which features of the case are explanatory primitives and which can be derived.

I defend an actuality requirement for a situated attempt: there must be an operative initiation or maintenance of a course toward the end. I leave open whether that actuality is always one separate event of trying, because it may be constituted by ordinary movements, by higher-order control, or by distributed performances. The decisive claim is that unrealized success alone neither creates nor removes the practical episode.

Practical Knowledge and the Epistemology of Trying

Agents can often know they are trying without observing whether the intended result has occurred. A person can know that they are attempting to call someone even while the connection has not been established. Such knowledge arises from the agent's undertaking and practical orientation. It may be available directly through the organization of action rather than through external inference.

Anscombe's account of practical knowledge makes this an epistemological issue. The knowledge characteristic of intentional action is not simply an observational report about one's body. It is connected with how the agent directs performance under a practical description (Anscombe, 1957/1963). Attempts sharpen the distinction because the agent may know the trying while being uncertain of the achievement.

This epistemic privilege has limits. An individual may believe they are attempting to restart a device when they have accidentally operated an unrelated control. The proper description of their practical episode depends on what their conduct is and how the equipment works. First-person authority concerning intention does not settle every fact about intentional execution. The person may know the intended end but misidentify the actual action.

There is also asymmetry between knowing that one intends and knowing that one is trying. An intention may be present while an agent remains inactive. The transition to trying sometimes depends on whether an action sequence has begun, whether a plan is in execution, and whether a relevant means has been taken up. These are facts that may require practical awareness, external feedback, or both.

Epistemic responsibility enters when agents assess failure. Someone who makes repeated attempts while ignoring clear evidence of a defective method may be acting, yet reasoning poorly. Conversely, an agent can recognize that the target outcome is impossible and rationally stop trying. The knowledge needed to evaluate an attempt includes information about available means, risks, changing circumstances, and the grounds for perseverance.

The philosophical payoff is a distinction between the ontology of an attempt and knowledge of that ontology. A person may genuinely try without accurately describing the attempt. A person may intend to try yet be mistaken that any attempt has begun. And an observer may know that an attempt occurred through the pattern of conduct and context even without access to the agent's experience. Action epistemology must account for all three possibilities.

Ethical Responsibility and Unsuccessful Agency

Moral evaluation regularly extends to attempts whose target outcomes never materialize. Someone who tries to prevent harm may deserve recognition even when the prevention fails; someone who deliberately attempts to harm may be blameworthy even if circumstances prevent injury. These judgments show that ethical appraisal cannot depend solely on actual consequences. It must include the agent's practical direction, reasons, choices, foreseeable risks, and control over conduct.

Kieran Setiya's treatment of practical reasons gives a further framework for asking whether a person had adequate reason to undertake an uncertain course and whether such reasons depend on the agent's beliefs and practical understanding (Setiya, 2007). A justified attempt can remain unsuccessful through contingencies external to the quality of deliberation.

The evaluation nevertheless differs between attempts and completed acts. The resulting harm, benefit, or irreversible change may matter morally in ways that intention alone does not. Discussions of moral luck illuminate this complication: how much of an agent's moral standing should depend on outcomes influenced by factors beyond the agent's control? Nagel's “Moral Luck” gives a classic statement of the problem, emphasizing tensions between control and judgments shaped by circumstance (Nagel, 1979).

Attempts invite an analysis with several dimensions. The agent's intended end matters; the means selected matter; the foreseeable probability and gravity of effects matter; the degree of initiation matters; and the agent's response to new information matters. A sincere rescue attempt undertaken with reckless disregard for bystanders may be morally mixed. A harmful attempt may be culpable despite its technical incompetence. These evaluations require attention to conduct rather than a single opposition between intention and result.

There is a further difference between blame for trying and blame for failing. An individual who undertakes a task beyond their competence may be responsible for accepting it, for the inadequate manner of trying, or for not seeking assistance. Yet when failure follows despite reasonable care, the negative result may provide no basis for attributing fault. Responsibility for outcome and responsibility for performance are conceptually distinguishable.

Legal classifications of attempt have their own institutional purposes, thresholds of commencement, and requirements of culpable intention. Their details vary across systems and should not be imported without jurisdiction-specific evidence. At the philosophical level, they illustrate an enduring challenge: society sometimes attributes responsibility before the prohibited result occurs, but must distinguish dangerous action from remote preparation and uncertain imagination.

An adequate ethical account therefore evaluates attempts as actual exercises of practical agency and situates them within broader normative contexts. The existence of an unsuccessful attempt can establish a proper object for praise or blame; it does not by itself settle the degree of either. Moral assessment requires an analysis of what the agent tried, how the attempt was performed, and what kinds of practical and moral control were available.

Collective and Technologically Mediated Attempts

Attempts are not confined to isolated bodies. A research team may try to reproduce a result, a medical team may try to stabilize a patient, and a group of engineers may try to restore a damaged network. These are candidates for collective action when individual contributions are coordinated through common plans, roles, and procedures. The relation between collective attempt and individual attempts is not always simple aggregation.

An individual researcher may successfully prepare a sample while the collective experiment fails. Another member may undertake a data-analysis task that contributes to the larger project without individually trying to complete the whole experiment. The team-level attribution depends on coordination and the practical organization of the work. A satisfactory theory needs to distinguish personal responsibilities, joint commitments, and the integrated causal process.

Technological mediation introduces further levels. Someone may attempt to move a robotic arm by transmitting instructions through an interface. If the remote mechanism is offline, the person may still have attempted the operation by issuing the command under appropriate control. The physical location of the effect and the physical location of the initiating act differ. The attempt is intelligible through the control channel, its designed causal role, and the agent's operative practical orientation.

The example helps separate trying from bodily exertion. A person may make a genuine attempt by operating a switch without unusual effort. Conversely, a technical system can exert considerable physical force without any human agent having undertaken that particular end. The proper unit of analysis depends on where practical direction and control are exercised.

Questions about artificial agents require careful treatment. Whether a system's goal-directed performance constitutes trying in a philosophical sense depends on the adopted account of agency: one must examine operational selection of ends, control organization, response to counterevidence, and the grounds for attributing reasons or commitments. Merely programming a routine to repeat an operation does not settle the issue. Equally, the absence of human phenomenology does not by itself resolve every question about functional descriptions of artificial activity.

The broader Artificial Era raises an important research horizon concerning technologically and nonbiologically realized action, yet the present dossier is concerned primarily with the general logic of attempting. It draws no new canonical definition of Artificial or Artificial Sapiens from the thought experiments considered here. The conceptual lesson is that attempts can be mediated by multiple channels and coordinated structures, provided the attribution of practical orientation and control is argued at the appropriate level.

Critical Cases and Counterexamples

Any proposed definition must withstand cases in which intention, conduct, success, and control diverge. These cases expose the limits of a simple formula and indicate how the concept should be refined.

The first case is pure intention without undertaking. A person firmly intends to apologize tomorrow but does nothing toward the apology today. The person has an intention, perhaps a plan, and perhaps a sincere commitment, but ordinarily has not yet attempted the apology. A theory equating every intention with trying makes the concept incapable of explaining commencement.

The second case is incompetent action. A person attempts to light a lamp by repeatedly pressing the wrong switch. The behavior is organized under a false instrumental belief. If actual causal efficacy were required for an attempt, the person's failure of knowledge would imply an absence of trying. That conclusion is implausible: mistaken means can still constitute an attempt, subject to limits concerning intelligible relevance to the end.

The third case is blocked bodily execution. A person tries to lift an immobilized arm. One cannot simply infer the presence of a successful bodily movement. The case asks whether motor initiation, volition, or the exercise of a practical capacity can ground trying without the intended movement. It remains a hard case for theories that require overt completion of any constituent act.

The fourth case is coincidence. A person attempts to open a jammed door and another individual opens it from the opposite side. The intended state occurs, but not through the attempted causal route. The success condition for the first agent's opening is not established by a merely matching result. Attempts require more than the concurrence of intention and outcome.

The fifth case is coerced participation. A person performs movements directed by another under severe external constraint. The person may still be acting in a limited practical sense while lacking broader freedom or responsibility. The account must distinguish the existence of attempting from the moral quality of the circumstances under which the attempt occurs.

The sixth case is repeated failure. A person tries a code several times before unlocking a device. The series may comprise several attempts at entry and one extended endeavor. The number of attempts depends on the relevant task boundaries, resets, and practical commitments rather than on a fixed rule that each physical operation is an independent attempt.

The seventh case is withdrawal. A climber begins attempting a route, recognizes imminent danger, and retreats. The failed ascent may involve a successful exercise of practical judgment. The account must not treat failure at the original end as an all-things-considered failure of agency.

The eighth case is accidental motion under an intention. The individual wants to activate an alarm, becomes frightened, and accidentally strikes the switch. Intention and causal influence may exist without a controlled attempt through that movement. The case justifies a requirement of practically organized conduct rather than a bare causal chain from mental state to outcome.

These cases collectively favor a relational and context-sensitive analysis. They also caution against one-factor definitions based solely on desire, intention, muscle movement, felt exertion, or achieved consequences.

The Structure of Attempts A Philosophical Synthesis

The analysis supports a five-part structure. An attempt has an agentive bearer, a target end, an operative orientation, an undertaken course, and a relation of possible realization. The bearer is the individual or adequately organized collective to which the undertaking is attributable. The target supplies the success condition. The operative orientation explains why these activities are directed toward that target. The undertaken course is what the agent initiates or sustains. The relation of possible realization links the course to the target through actual or believed means.

The structure can be represented schematically as A attempts E through C at time T when A practically directs and initiates or sustains C toward E at T, and C is intelligible within A's operative understanding as a way of pursuing E. The representation is a conceptual guide. It does not claim that an intention must be consciously verbalized, that the means must be effective, or that success must be genuinely possible in an objective sense. These stronger conditions would eliminate ordinary attempts made under mistaken beliefs.

The undertaking condition is decisive. A person can wish, deliberate, resolve, or plan without trying. Attempt begins with a practical transition into execution, understood relative to the level of the action being described. In some cases that transition is a bodily motion; in others, a command, a coordinated group act, a motor initiation, or an active process of inquiry. There is no requirement that all attempts have a single uniform physiological signature.

The means-end relation is likewise essential. Random behavior cannot become an attempt merely because it coincides with a desired outcome. The end must organize the conduct through an agentively intelligible route. The route may be mistaken, improvised, or blocked, but it must form part of the agent's practical orientation. This requirement answers the problem of causal deviance by locating the attempt in appropriately guided performance.

Success introduces an additional relation. The attempt succeeds at E when E is realized through the relevant directed course in the manner required by the action description. The agent may have succeeded at constituting subactions without realizing E, and may have achieved an E-like state through coincidence without successfully doing E. These distinctions secure the conceptual integrity of both trying and achievement.

This synthesis does not determine whether every attempt must be classified as an independent event token. The positive undertaking may itself be a bodily event, an integrated process, or an organized sequence. A metaphysical theory may identify a distinct trying occurrence; another may analyze the attribution through activities and counterfactual relations. Both must account for the same practical structure and hard cases. The inquiry into the ontology of trying remains a genuine philosophical dispute after the structure of attempting is clarified.

The historical contribution of the analysis is equally important. Aristotle supplies voluntariness and choice; early modern philosophy separates willing from effective bodily power; Wittgenstein challenges a facile inner addition to movement; Anscombe identifies practical descriptions and practical knowledge; Davidson connects reasons and causation; Frankfurt foregrounds ongoing guidance; O'Shaughnessy and McCann place volitional exercises at issue; Bratman and Mele distinguish plans and intentions; Ruben and Payton dispute the truthmakers of attempts. Their positions are neither interchangeable nor reducible to one current vocabulary, but they jointly illuminate the principal dimensions of the concept.

The conceptual result is that attempts are the form in which agency remains describable across the gap between practical initiation and uncertain realization. They make visible the causal exposure, temporal unfolding, and normative assessability of action in a world that need not comply with an agent's purposes.

Open Philosophical Questions

Several questions remain open after the preceding reconstruction. The first concerns ontological identity. Must each genuine attempt contain a particular event or process, or can some attempts be made true by an agent's powers, orientation, and appropriately restricted counterfactuals? The challenge is most acute in cases of paralysis and complete external blockage, where the evidence for an occurrence is difficult to distinguish from evidence for an unexecuted intention.

The second concerns the threshold of initiation. When does preparation become trying? A student who enrolls in a course may be trying to qualify for a profession, but may be merely preparing for a future attempt at a specific examination. The threshold is sensitive to practical level and social convention. A more precise account may need separate criteria for basic bodily attempts, institutionally defined attempts, and temporally extended projects.

The third concerns the individuation of repeated effort. Do efforts separated by interruption remain one attempt if unified by a plan? Is the distinction between an attempt and an endeavor partly pragmatic, depending on which successes and failures matter for assessment? A full theory must explain why some nested or repeated episodes are naturally counted independently and others are not.

The fourth concerns the role of control in atypical cases. How much guidance is required for an occurrence to qualify as an attempt, and how does control differ across reflexive, skilled, deliberate, and technologically mediated performance? An overly stringent requirement excludes ordinary fallible activity. An overly weak one admits causal coincidences.

The fifth concerns practical knowledge. Can an agent be mistaken about whether they have started trying, while correctly understanding their intention? How should first-person authority be reconciled with mistaken instrumental beliefs and misleading feedback? The answer affects theories of action knowledge and self-understanding.

The sixth concerns nonhuman and collective agency. Under what conditions can a group or an artificial system attempt an outcome as an agent, rather than serve as an instrument of someone else's attempt? The answer will depend on substantive theories of agentive organization and practical direction, and cannot be settled through verbal analogy.

Finally, ethical theory must explain how moral evaluation tracks attempted ends, the manner of undertaking, and contingently realized effects. Failed attempts reveal that a moral theory concerned only with intentions misses actual conduct, while one concerned only with consequences misses the directed history of agents. The open question is how these dimensions should be weighted across different normative contexts.

Conclusion

Attempts belong to the fundamental grammar and structure of action. A person who unsuccessfully tries to do something may nevertheless have acted, because the reality of practical engagement is established by what the person undertook and how that undertaking was directed, rather than by the eventual completion of the intended end. This conclusion is compatible with important disagreements about the precise ontology of trying.

The historical reconstruction shows how the issue develops across distinct philosophical programs. Aristotle's analyses of voluntariness and choice identify the practical source of conduct. Early modern accounts of willing separate purposive direction from effective bodily power. Wittgenstein reveals the inadequacy of a simple inner-act supplement to movement. Anscombe and Davidson make practical description, reasons, and causation central to action explanation. Frankfurt, O'Shaughnessy, McCann, Bratman, Mele, Ruben, and Payton develop competing accounts of guidance, volition, planning, and the metaphysical conditions of trying.

The monograph's conceptual synthesis identifies five necessary dimensions for analysis: bearer, end, operative orientation, undertaking, and means-end relation. No one dimension can do the work of all the others. Intention without initiation can remain a plan; bodily movement without practical orientation can remain an accident; felt effort without a relevant undertaking can remain a sensation; the occurrence of a desired outcome without an appropriate action-guided route can remain a coincidence.

Failures are particularly informative because they expose these separations. A defective mechanism can prevent an end while an agent acts purposefully. A blocked motor system can make the ontology of agency difficult to locate. Extended projects can persist through repeated local failures. Collective efforts can involve successful individual contributions without achieving the team's goal. In each case, the philosophical task is to describe accurately what was attempted, how it was undertaken, and which of its success conditions were met.

The final result is a general account of attempts as real, situated, practically directed engagement under conditions of contingent realization. This account makes it possible to investigate knowledge, skill, responsibility, and nonhuman agency without equating action with accomplishment. Its remaining ontological questions are precise: what constitutes undertaking in the difficult cases, when attempts begin and end, and whether a separate event of trying is indispensable. These questions define the continuing philosophical research program.